They are always available, and always responsive. Their green dot on Slack never disappears. They give you thirty options when you only needed three. Their calendar is full, but they keep working after everyone else has stopped.

Are they good at their job?

Perhaps.

They might be exceptional. Their force of will might be the only thing keeping an organization together. Carrying impossible work that their manager doesn’t even understand, and their colleagues never see.

They might be completely overwhelmed. Inefficient. Anxious. They might be producing a stream of work that makes more work for all the rest of us.

And the midnight email doesn’t help us know which one.

We know that manager who mistakes dedication for performance.

They see the hours at your desk. They look at the messages you send, the meetings you attend. The tickets closed and the lines of code written. They count the documents you produce.

Did that message change an outcome? Was that outcome worth the effort?

Now we have AI. We’ve built the employee who sends emails at midnight.

We haven’t changed its manager.

It’s permanently available. It never gets tired. It responds immediately, every time. It can outmatch any human worker in its capacity to generate reports, plans, code, images, summaries, strategies - every document or artifact you could ever want.

If we’re measuring by presence, effort, and production then AI is the greatest employee who ever lived.

That’s probably not helpful.

The things a manager sees

When I was leading the Carbon Design System at IBM, I brought adoption targets to Phil Gilbert, the GM of IBM Design. He told me to stop focusing on adoption. It wasn’t the right measure of success.

Adoption was interesting. It wasn’t inherently valuable.

That conversation was years before we started asking the same question about a machine.

There are six different questions we can ask about work.

Presence: Was someone at their desk?

Effort: Did they put in the work?

Production: Did their work create outputs?

Performance: Were those outputs any good?

Outcome: Did they change anything?

Value: Was the change worth it?

Good management, good measurement, will look at all six but judge mostly through the final three.

Presence and effort matter. We like people who show up. And we usually want work to produce something.

Those things are necessary. But they don’t demonstrate success.

Performance is an assessment of the quality of the effort, not just the volume. Outcome lets us ask what changed in the world because the work was done. Value asks whether that change was worth the cost and risk.

Those are harder measures. At a minimum, they need someone to define what “good” means.

Bad management - maybe even average management - reverses that importance.

The employee who stays late looks more committed than the employee who finished the important work and left at 4pm. The person who sent you a fifty-page document looks more substantial than the one who sent you a two-line email that identified the decision that mattered. One team closes a hundred tickets, and looks productive. Another team prevents the creation of a hundred tickets, and looks quiet.

And it’s not just managerial foolishness. It’s not malice.

We understand bums in seats. We understand people working late. We understand the production of stuff. It’s all trackable. It all shows willing.

Value emerges more slowly. There are dependencies. More people are involved. Value might be stopping some work from ever happening. It needs an understanding of quality.

And we’re not great at that nuance. So organizations reward the visible. Employees learn to be visible.

If you’ve sent an email after 6pm when you could have sent it before 6pm, you know what I mean.

That’s about showing that you’re working. Not about showing that the work matters.

The proxies used to mean something

We valued presence, effort, and production because they’re connected with human limits.

You have to be there. You get tired. You spend your own time making stuff.

If you’re a person who made twenty substantial reports, beautifully presented as PDFs, you’ve done a bunch of work. That doesn’t mean the reports are good, but shows you’re invested.

Stuff that takes time, that takes effort, has scarcity.

Until AI came along.

AI can be there every hour of the day. Its effort doesn’t come with fatigue. It can make a lengthy report while you’re making your coffee.

Human limits secured that proxy.

Now? Delivery is decoupled from investment.

AI isn’t padding its hours to optimize a number. It genuinely is always available, always responsive. It’s making a lot of stuff. Nothing is being faked. The activity is real, and it’s honest.

But whether it’s valuable...that’s another question.

Here’s the kind of evidence we’ve seen for AI success:

• The system generated ten thousand lines of code.

• The system answered five hundred questions.

• The system created forty campaign concepts.

• The system summarized every customer interview.

These might all be true.

• Was the code reliable, maintainable, or necessary?

• Were the answers useful? Correct? Did anyone act on them?

• Were the forty concepts any better than the three that the team developed without that tool?

• Were the summaries accurate? Did they reflect what was really important to users?

They aren’t, by themselves, value claims. They’re diagnostics.

Is the machine operating?

The industry has automated its weakest managerial instincts and called it “observability”.

Production is not productivity

Generative AI is seductive. It makes stuff.

Blank page to full page. Empty backlog to populated. Busy inbox to drafted replies.

It did something. And we, as humans, like something more than nothing.

And we’re still adjusting to the sense of movement.

Before we had generative AI, making that report was - at a minimum - demonstration of effort: research, organization, and writing. The size of the artifact was evidence that we’d worked hard.

AI completely breaks that relationship.

That long report might come from a one-line prompt. It might never provide value.

The artifact, when it had human cost, looked significant. AI strips that cost. But as humans we still respond to the length, fluency, and completeness. We want the work to be consequential.

The AI stayed late. Look at everything it made.

AI gives you execution speed, no question. The correction tax is an efficiency cost.

Who’s going to read the report? Which developer will be reviewing the code? Am I going to have to choose between the fifty options it gave me? I’ll need to check the citations. I’ll go through the error logs to find the gaps in a convincing answer.

That isn’t a productivity increase. It’s an increase in paperwork.

We don’t demo the cost part of AI, just the speed.

The work management avoided defining

I wrote recently about auditing the AI skills I had built for my own work and life. The audit showed me that measuring the machine was just the beginning.

We only know if the AI-generated product brief is good if we actually know what a good product brief is. To evaluate a coding agent, we need to decide if we care about delivery time and defects, or lines written.

We can’t rely on “I’ll know if it’s good if I see it.” That’s a managerial dodge.

And it’s not about one universal definition of quality.

We need stated definitions for success, quality, and value, in the context the work is delivered. And those definitions should be owned by somebody willing to defend them.

Organizations have left this vague for years. Effort is reassuring. Good managers - able to apply their contextual understanding - filled gaps with their own judgment on impact.

AI makes those gaps harder to ignore.

AI can generate more activity than we can inspect. Faster to production also means faster to shipping mistakes.

If production is unlimited, what does production tell us?

This is why evals matter.

It’s not about reducing every aspect of work to a score. Quantitative measurement doesn’t remove judgment.

Who receives the benefit, and who absorbs the cost? A serious eval makes those kinds of judgments explicit.

Evals shouldn’t just be tests for the AI systems.

They’re tests of whether management knows what it is asking the system to accomplish.

The performance review

AI looks amazing under the weakest measures of productivity.

It is always at its desk. It works with astonishing speed. It’s never tired. It produces more than anyone else.

Perhaps it is also doing excellent work. Perhaps it is changing outcomes that matter and creating value far beyond its cost.

But the first three things don’t prove the final three.

The midnight email never proved value.

Evals move the performance review beyond production.

It means defining what good looks like. Being able to sift through volume for value. Not simply admiring the output.

AI worked through the night. We already know that about AI.

We need to know whether the work matters.

Further reading:

• Design system adoption numbers...just a vanity metric? - the same argument, before AI: adoption is interesting, not valuable.

• I know kung fu. I might remember how to throw one punch. - the skills audit referenced above.

• Collins, S. We Hired AI to Work Less. Instead, Our Workload Jumped 346%. Here’s Why. Activated Thinker, Mar 2026.

• Schawbel, D. How AI turned your best work into the bare minimum. Fortune, Aug 2026.

Article photo by Bruno BD on Unsplash.

Solomon heard it first, the ugly cawing of a gull. “There!” he croaked, pointing at a speck on the horizon. “Must be land nearby.”

Two more hours of drifting before we spied a dim smudge above the line of blue ocean. Hope was enough tonic to give us strength to paddle with our hands. The energy to creep towards the distant isle, and possible salvation.

I awoke later, who knows how many hours. Andrea was shaking my shoulder violently. “We’re in the shallows,” she hissed. “Boats breaking up, gotta swim the last.” She pointed to the stony beach ahead, a few hundred yards.

Any one of us alone would’ve drowned, I’m sure. But together, somehow, we paddled, and floated, and dragged each other forward. I was completely spent as I felt my feet touch rocks, and I could barely take the last few steps to cast myself upon the beach. My companions collapsed beside me.

The sun was low when I next opened my eyes. Every inch of me ached, but I rose and nudged Andrea and Solomon to wakefulness. “We need water, and a fire, before the sun goes down.”

Inland a few hundred yards we came upon a small stream, and the fresh water tasted like nectar. Solomon searched the beach and found enough driftwood for at least a few hours of fire. As Andrea knelt to set the kindling, I wandered in the hope of finding food.

Instead, in the dim light, I stepped in a hollow and near broke my ankle. Cursing in pain, I looked down, and was shocked to see a glint of metal. Examining more closely, I made out the corner of a metal-banded chest, poking out of the ground. Something buried carelessly. Left exposed once the elements had eroded the earth that covered it.

The light was near gone, now. I was lucky that Andrea had sparked a fire that lit the night, a beacon for me to return. I told my friends my discovery, and we determined to investigate the next day. Curled up on grass and sand, we slept like the dead.

Solomon gestured to me to pull on the corner of the chest. He levered underneath with a plank of wood, a remnant of our broken boat. Andrea shoveled sandy earth away with her hands until, with a groan, the chest finally released itself. All three of us fell back, exhausted from the effort.

The lock was rusted, and broke easily under my heel. And our eyes reflected the glow of gold and silver as we cast open the lid. “More’n the crew’s seen in years,” said Andrea.

“Aye. An’ much good it does us,” said Solomon. “We can’t eat gold.”

“We find some more wood,” I said. “Stoke the fire. Forage. Survive until someone finds us.”

“An’ then they just carry us off to a friendly port, and give us all the gold, instead of cuttin’ our throats and takin’ it themselves?” asked Andrea.

“They leave some of it for us. If they don’t know it’s there. Show any rescuer this still full chest we found, and cry tears of gratitude. I’ll wager they don’t ask many questions.” I smiled, “And in the meantime, we’ve gold in our belts, in our boots, sewn into our clothes. Enough to be rich, an’ still alive.”

Solomon nodded. “Aye, ‘t might work.” He looked at Andrea, who nodded, too.

That was a week ago. The fire hasn’t gone out in that time. And we’ve survived on shellfish and a few roots. My stomach still growls. But there’s a chink of metal when I go to tighten my belt.

And there’s a sail on the horizon.

Article photo by Patti Black on Unsplash.

Should this person be allowed to generate that?

That’s identity management and role-based access controls. Audit logs. Data-retention policies. Spend limits (now that we’re realizing tokens aren’t free!). Permissions for tools and connectors. And controls over the models available, what information they can reach, and what gets recorded.

This is good. Sensible and secure.

It’s also only half the problem.

There’s another question that gets missed.

Is what gets generated acceptable?

We’re still talking about governance. But the first question was about access. The second is about conformance.

Enterprise AI stacks are much better at the first one than the second.

Anthropic has built in SSO, SCIM, role-based permissions, audit and compliance tools, data-retention controls, observability, spend management, and controls to configure how models connect to outside systems. Microsoft has a growing governance layer around Copilot. Okta is putting work into expanding identity governance so it captures agents as well as humans.

Enterprise companies know how to solve these problems.

I’ve worked at IBM and at J.P. Morgan. These are the types of companies who need to determine who someone is, what they can access, and what they did.

So it’s natural that they apply the same methods to generative AI.

Who invoked that model? Were they allowed to? What data and tools could it use? How much did it cost?

And these could all work perfectly.

The authorized employee signs into the approved AI product at their company. They have all the right permissions. The model is locked down - only the information they’re allowed to see. The interaction is logged, there are traces. Nothing leaks, no policy is breached.

The AI makes some kind of plausible artifact.

And it’s wrong.

Maybe not obviously. But wrong according to all of the organization’s own standards.

It violates accessibility. It ignores a code convention. It’s in conflict with an existing product decision. It invents a component that’s already in your design system.

Nobody catches this. It looks right, and it passed all the governance controls.

It ships.

The governance stack only solved access. It protected the company from misuse of AI.

Conformance governance would protect it from AI that makes the wrong thing.

We’re building the first one faster than the second.

Not my first governance rodeo

Before AI (and after), I’ve spent years working on design systems.

Design systems always had a governance problem - more than they had a component problem.

The component library is the simpler part. The hard questions come when hundreds or thousands of people start using it.

Introduce standards and training. Establish review procedures and processes.

And, if you’re not careful, you become the design systems police.

I wrote about that problem four years ago.

That central team discovers their job isn’t helping people make good decisions. It’s turned into catching people who make bad ones. It’s not governance, it’s enforcement. The system is a gate.

That wasn’t scalable when the people doing the work were all human.

AI’s made the same problem much larger.

A designer can create five credible options in the time it used to take them to make one. Now there are five times as many things that need checking.

Same on the engineering side. There’s much more code to review.

Did the ten pages of analysis that the product manager made this morning align with what the company knows?

The smaller the generation problem, the bigger the output problem.

And we’re only just starting to pay attention to that part of the productivity story.

We’re all striving for velocity. For me, that means a combination of speed and quality.

AI can reduce the cost of making an artifact from four hours to four minutes.

Then we have to spend thirty minutes reviewing that artifact. Another thirty to identify all the deviations. An hour of corrective work. Coordination with other people who also need to review.

The work’s moved, not disappeared.

AI without conformance means the generation cost is converted into correction cost.

And right now more of the correction remains human. So the faster generation makes things even worse.

I wrote earlier this year that execution is becoming cheap while coordination is not.

This is one of the consequences.

The bottleneck moves downstream.

System discovery isn’t the problem

In early evals work at Knapsack, we started to see a clear distinction.

A controlled evaluation. Test a coding agent doing design system tasks. A matched task set, no customer data. Give it two conditions. One where it has an MCP connection that provides guidance for that system. One where it doesn’t.

The MCP doesn’t help the agent find the system. It does that successfully, assuming it has access. It even imports the components at about the same rate.

The agent knows to look in node_modules.

But if it doesn’t have the MCP serving the context, it doesn’t know how to use that system.

Give it that context, and design fidelity improves 5-10%. Code quality improves by more than 20%.

Prompts that produced shippable code rose from 40% to over 60%.

TypeScript errors down significantly. Prop violations similarly reduced.

The cost of running any individual task was higher. But the cost per shippable output was thirty percent cheaper.

This agent didn’t have new access. It already had the components.

But now it had context about how the organization expected the components to be used. Props, variants, compositions, and the constraints around them.

The access problem was already solved. The conformance problem wasn’t.

Conformance improves the economics of usage. Even with our early, simple context provision the benefits are clear and measurable.

The organization’s intent becomes a participant in the generation.

Plausible mistakes compound

The more complex the task, the more the agents benefit from additional context.

On a multi-screen flow, attaching the MCP gave us gains that were nearly double what we saw on simple patterns.

This isn’t a neat scaling. We don’t have enough data. On some simpler template-based tasks, the additional context sees the agent second guess itself and reduces its reuse of the right components.

But high-complexity tasks, broadly, benefit the most.

As tasks get bigger, they involve more decisions. And multiple small plausible but wrong decisions start to compound into something globally wrong.

I always talked about it when it came to design systems. You can take a bunch of completely accessible components and assemble them into a really inaccessible experience.

A complete flow can violate how your company believes the experience should work.

Conformance is broader than compliance

“Conformance” isn’t just another word for regulatory compliance.

Conformance is a broader evaluation of whether an artifact satisfies the constraints your organization has already decided.

Yes, be compliant with accessibility. Dealing with private information. Legal restrictions your company needs.

But some of the constraints and context are just how the organization has chosen to operate.

Brand standards.

Design-system conventions.

Product strategy.

Architectural decisions.

Content guidelines.

Approved patterns.

These are things that teams throughout your company have learned painfully before. And they’d prefer not to learn them painfully again.

Any organization is full of these decisions.

But they’re scattered.

Design system sites. Wikis. Policy documents. An old presentation. Architecture records. Jira tickets. Multiple Slack threads. That developer’s memory. The shared knowledge of the people who were in the room when some decision was made.

AI isn’t magically better than humans at pulling all this context together when it’s time to execute.

Given incomplete organizational context, it’s probably worse. It’ll make more guesses, ask fewer questions, and generate something plausible from the context it does have.

Which is what we asked it to do.

And generic AI evaluation only gets us to a certain point.

A benchmark can tell me if Opus 5 is generally good at coding.

An eval can tell me if GPT 5.6 tends to complete some specific task successfully.

It doesn’t tell me if the code it’s producing is what IBM wants to ship. Or if the artifacts it creates follow Amazon’s design standards.

I say it a lot. AI is very good at making “plausible but wrong.”

Conformance is contextual to the environment.

But we can’t just make another gate

The default enterprise solution is probably going to be simple.

Review everything.

Create an approval process.

Put a human in the loop (...that poor human).

Build a better police force.

It won’t work.

Generative systems are supposed to dramatically increase the amount of plausible work produced. We couldn’t scale human review enough when humans were the only ones generating work. It’s even more impossible to grow proportionally with AI generated output.

Which defeats the economics of using that system to govern AI.

The best design system teams I’ve worked on weren’t successful because they were really efficient at rejecting bad work.

They made doing the right thing the path of least resistance.

Documented decisions. Context. Clear constraints.

Visibility to what already exists.

People were guided towards conformance while they worked. It wasn’t a case of coming to the end of the day and realizing they’d violated some rule they never knew about.

I’ve described this, slightly in jest, as a virtuous panopticon. Not watching everyone. Instead making the possible and the preferred more visible.

The same applies to AI. Conformance can’t come after, it needs to participate in the generation.

Guide the way, don’t gate the path

So that’s an interesting new enterprise AI infrastructure problem.

Not how to restrict what models can reach.

How do we make our organizational standards available to them while they work.

If the model is producing UI it needs to know components, accessibility requirements, voice and tone, patterns, and interaction conventions.

If it’s also writing code, it needs access to architecture decisions, security policies, dependency rules and engineering conventions.

Output needs to be evaluated against those constraints.

Block some failures before they’re made. Warn about others. Point to an alternative.

Or just make the relevant context more visible to the human-in-the-loop.

Governance can mean guidance as much as it means permission.

It’s not enough to have standards. It’s about having standards that are easy to follow.

Which, if you’ve worked in enterprise, you’ll know is very often not the case.

There’s a standard. Or six.

Policies and approved patterns.

And the people doing the work don’t know where to find it. Or how to choose which one they should use.

If it’s a human, they probably stop and ask someone.

Generative AI doesn’t stop. It produces some finished-looking artifact.

That can be a hard failure to notice.

The missing half

AI access governance needs to be sophisticated.

We need all that information about users, controls, audits, permissions and boundaries. Especially when we’re monitoring autonomous systems.

But those controls don’t tell us whether the thing those systems produced should exist.

That’s the conformance layer.

The layer that goes beyond just who can ask the question, but whether the answer that comes back is right.

Approval queues don’t work as governance at scale. They slow everyone down. They make people hate the expert teams reduced to policing them.

AI can actually give us an opportunity to make it better.

Make our standards more easily available when the work is being created.

That means encoding organizational knowledge at the right level. And evaluating our outputs directly against what the organization has already decided.

Guide people and agents in the right direction, before we have to stop them going in the wrong one.

Access protects the company from the AI. Conformance protects the company from what it asked the AI to do.

Further reading:

• Don’t become the design systems police - on showing up as a facilitator and not a blocker.

• Nashawaty, P. & Weston, S. AI Output Governance: The Blind Spot in Enterprise AI Strategy. Efficiently Connected, Aug 2026.

• Sure, R. W. The Enterprise AI Governance Layer as a Control Plane for Trusted Enterprise Intelligence (pdf). Independent Research Paper, Jun 2026.

• Mugel, S. Enterprise AI’s Governance Gap: Runtime Safety Is The Missing Layer. Forbes, Aug 2026.

Article photo by Annie Spratt on Unsplash.

Maybe a seed came off a tire tread. Or someone’s foot. Or blew there in the wind.

It fell into a crack in the concrete outside the ramen shop. It was the second week of March.

By April there was a stem.

Three floors up, behind glass, a NutriSync Cultivate unit ran its first cycle.

Welcome to Cultivate.

Your unit is now delivering 96.2% of ideal photosynthetic conditions. Spectrum, humidity and nutrition dosing are managed for you.

Sit back. Harvest will come sooner than you think.

On the street there was no spectrum. Just a gap between two buildings, and the sun that got through in the afternoon. The plant leaned toward the sun.

Feet passed. In May someone stood on the second leaf, and the stem laid flat all day. Then it got up.

In June, the street cleaning drones ran a scheduled service at 4am. They missed the crack in the concrete by an inch.

Upstairs, the tray stayed at 96.2% of ideal conditions. It produced a pale seedling that barely outlived its first month. It got thrown into the trash, planting tray and potting medium together.

The plant grew wilder in August.

It flowered in late summer. Then it set seed. In November, frost came off the water and it shivered. Browned, but stayed standing. Its seed heads were still intact in December.

Sit back. Harvest will come sooner than you think.

Somewhere down the port corridor, in a joint where the concrete had moved.

A story from the Static Drift universe.

Article photo by Lala Azizli on Unsplash.

Give a designer and a developer the same product brief.

Give them the same amount of time. The same access to users. The same AI tools.

The designer needs enough technical awareness to understand that architecture, performance, accessibility, and security are real constraints. The developer needs the design awareness to recognize familiar interaction patterns and make a coherent interface.

Let them work independently.

At the end of the time, don’t judge the codebase. Don’t count the features. Don’t ask which interface is more polished.

Put the products in front of users and see which one better solves their problem.

My bet is on the designer.

That wouldn’t have been the case a few years ago.

The gates were never symmetrical

Designers understand users. They can frame a problem. Structure the information they have into useful conclusions. Take all that, and determine what the experience should be.

But they couldn’t ship it.

They could make screens. They could have a click-through Figma prototype of the interactions. Explain the behavior with notes, annotations, tickets, meetings, and the all-important handoff.

So that someone else could make it real.

Developers had an opposite advantage. They can take an idea and make a working piece of software. What’s behind the interface - dependencies, data, performance implications, and how simple requirements get complicated very quickly when they meet reality.

But they didn’t have the judgement to decide how that software should work for a person.

Obviously these aren’t universal limitations. I know plenty of great designers who can code, and developers who have excellent design judgment.

But disciplines are training. They point our attention, so we notice different things.

Design trains people to see what’s confusing. What is incorrectly emphasized. The breaks between what a system allows and what a person is actually trying to do.

Engineers are trained to see fragility and bad abstractions. Understand the architecture and hidden dependencies. Bridge the gap between a convincing demo and a system to survive production.

Both those kinds of judgement are important.

AI hasn’t affected them equally.

Didn’t we always want designers who code?

AI gives the technically aware designer a remarkable amount of implementation capacity.

Scaffold an app. Connect it to APIs. Generate components or use the ones that exist. Explain unfamiliar code. Debug when things go wrong. Write a test suite. Take a clear spec around intended behavior and execute on it.

It will still need direction. A designer will need some technical understanding - enough to recognize if the system is making dangerous assumptions. To know if something is moving beyond their competence.

But it’s not the hard stop it used to be.

Designers don’t have to persuade a production chain to make something in order to discover whether it works.

They can make it.

AI also gives developers greater access to design production.

It can generate a clean dashboard. Use a familiar onboarding flow, or a plausible settings screen. It knows visible conventions of software very well. Sensible spacing, tidy cards, useful empty states - everything to make a product feel like it’s finished.

These upgrades aren’t symmetrical.

The system can increasingly perform implementation on the designer’s behalf. But design judgement can’t be acquired by a developer asking a system to produce design artifacts.

The artifact isn’t as valuable as the judgement. As the critique. As the understanding of the user.

AI passes the “first look” test

AI can make some very plausible interfaces. Many product generation tools lead with that capability. Here’s an immediate, visual, impressive interface.

The product appears before your eyes.

That looks like the design was the easy part.

It’s evidence of something else.

The presentation layer is the part of software applications where it’s easiest to manufacture the appearance of correctness, and hardest to verify if it’s actually correct.

If code is plausible but wrong, there are usually backstops. It doesn’t compile. The tests fail. There’s an error.

Engineering is great at detecting what’s incorrect because - if they don’t - the machine doesn’t execute its instructions.

Plausible-but-wrong design is dangerous because it can work perfectly.

The interface renders. The button works. The form submits cleanly.

It’s a great demo.

And there’s no alarm because the dashboard surfaced the wrong metric, and hid the one the user needed to make a decision.

There isn’t the same binary test when we cleanly and efficiently guide our user to the wrong outcome.

If we make a destructive action overly convenient, the app still builds.

We’ve got something usable.

We’ve got something wrong.

Those costs will come later - after it ships. Abandoned tasks. More support tickets. Bad business decisions because the information was wrong. Seemingly inexplicable churn.

It’s because AI is so fluent in the generalities of finished software that discrimination becomes more valuable, not less.

It’s not a matter of taste

This isn’t domain protection.

I’ve led design systems. Of course I think designers are important.

But the argument only works if design judgment means something tangible.

It’s not about choosing better typography, or adding visual polish. It’s not “make the logo bigger”. If it is, then AI has eaten up most of it already.

Some design execution can be automated - because production work in every discipline is getting easier and easier to automate.

Designers don’t have better taste.

Designers are best trained to determine if an experience represents what the user actually wants and needs to do.

What decision is the user trying to make? What information changes their decision? Which actions are reversible?

Does the structure of the product match the structure of the problem?

That’s judgement rather than taste. And, while not exclusive to designers, it’s what their discipline is supposed to train.

The design-aware developer knows that a destructive action may need confirmation.

The designer asks why the destruction is even an option.

It’s different below the waterline

The argument works on the surface. At the level the user sees.

Deeper than the presentation layer, and the results invert quickly.

Ask those same two people to create an infrastructure for high-scale transactions. To protect sensitive data. Recover from partial system failure. Avoid making an architectural decision that will constrain the company for a decade.

Now it’s the engineer’s judgement that’s the scarce thing.

AI produces plausible architecture, too.

So it’s not that designers now outrank developers. That’s dumb. And it reproduces the same mistake companies always seem to make - taking multidisciplinary product development and trying to turn it into a contest between the disciplines.

But AI shifts capability based on the nature of the missing skill.

AI is good at execution.

AI is bad at judgement.

If your historical limitation was execution, then AI gives you more benefit than the person whose historical limitation was judgement.

At the interface layer, that person is probably a designer.

Our org charts were built for the world before AI

Companies are making headcount decisions right now.

They’re compressing design teams, and ensuring engineering is the presumed center of their product creation.

And there’s plenty of logic to that. Hard engineering problems still exist. Software needs to work. And their AI investment is framed around making developers faster.

But that assumes the old distribution of capabilities.

The ability to ship is the decisive gate. Whole organizations have formed around that gate.

Now more people can go through it.

A designer who understands enough about software can increasingly go through it. Move from problem framing to working product, without a bunch of translation layers.

The reverse gate isn’t widening in the same way.

A developer can ask AI to produce a convincing interface. They can’t easily tell if that interface has the right understanding of the person who’ll be using it.

AI won’t be reliable at warning them.

Companies cutting design while concentrating on engineering investment “because AI can do the UX” might be reinforcing a discipline whose historical advantage is eroding the fastest. And reducing or removing the discipline whose central value AI is least suited to reproduce.

AI has made it much easier for designers to become builders.

It has not made it equally easy for builders to become designers.

Further reading:

• Execution is cheap. Coordination is not. - on catching what’s organizationally wrong and codifying it.

• The Intersection of Design Thinking and AI: Enhancing Innovation. IDEO U, Jun 2025.

• Hill, M. Half of developers say AI can code better than most people. AI Data & Analytics Network, Aug 2025.

• Morton, P. Why is AI bad at design? Phil Morton, Jul 2026.

Article photo by Eyre June Bustamante on Unsplash.

Last week I audited my entire toolkit of AI skill suites. Telemetry had its own verdict.

I didn’t use the fitness suite. At all.

I’ll come back to that verdict.

In The Matrix, Neo jacks in, downloads a file, his eyes open: “I know kung fu.”

Using my deep research skill to define a strategic plan for a topic, and turning that into an AI skill suite, is the closest thing I’ve come to feeling like that for real. Research goes in, a tool comes out.

Capability acquired.

I’ve had that feeling multiple times. I have twenty-two domains for work and life - 193 skills in total. Product specs, competitive analysis, balcony gardening, wardrobe management.

Do I know kung fu? Or do I just have a big folder of downloads?

I’m a bad person to answer my own question

In 2025 METR ran a randomized trial. They gave experienced developers frontier AI tools for their own codebases, and measured what happened. The developers believed the AI tooling made them 20% faster.

They were 19% slower.

METR ran a follow-up in late 2025. The slowdown headline went away - newer tools, more practiced people. More interestingly, their measurement broke. Developers wouldn’t even submit tasks they’d have to do without AI. And time-tracking was unreliable when they applied it to agentic multitasking.

The people who measure this as their job concluded that their own results weren’t a good “proxy for the real productivity impact.”

So the durable finding isn’t the 19%. It’s the gap between what we feel and what we can verify - and that gap is getting harder to close, not easier.

This isn’t about developers, so much as it’s about our own testimony. Fluency feels fast. My tooling makes me feel capable, which makes me less reliable. With nearly two hundred tools, I’ve constructed myself into the least reliable witness possible.

So I decided to see if I could pull the data.

• Invocation - what actually ran, when, and where.

• Artifact trail - whether the tools produced anything, and whether that went anywhere further.

• Research lineage - what was fed downstream.

And, fair to say, the telemetry itself isn’t perfect. It’s definitely missed skills that I know I use. Capture is hard.

But the numbers are more honest than my own feelings...my own vibes.

What did the audit say?

In the window I measured, 23% of the skills I’ve built were used. And only twelve skills carry half of all the activity.

That sounds terrible! 193 skills built, and I only really use twelve of them. What a failure!

But I don’t read it that way.

That’s not failure. It’s portfolio selection.

The marginal cost of building the skills with AI was tiny. And when the cost is tiny, the rational strategy is “build a bunch of stuff, and then let reality select.”

While the cost of building might be negligible, the cost of not deciding what to do with it is not.

My sin wasn’t building too much. It was that it took me this long to run the curation.

Living, dead, and somewhere in-between

When I look across my whole portfolio, I can break it down into six different states.

• Regular use. My daily workhorses. A dozen skills doing half the work - meeting debriefs, deep research, spec writing. No notes, these are clearly things I actively reach for.

• Retired, with honors. My first delivery suite - twenty commands - I used heavily for months. But it’s silent because I built a successor and ran an A/B between them. Three of six predictions were wrong - including one I’d filed explicitly as “what would particularly surprise me.” The new suite won, the old one retired. That’s a system that works.

• Obsoleted by drift. I had sprint-cycle skills. We moved from cycles to Kanban. The tools weren’t wrong, they just didn’t apply any more.

• Dormant. I have some skills around tax preparation. They’ve done nothing all summer. That’s correct - they shouldn’t be doing anything in summer.

• Graduated. An interesting one. My wardrobe and style suite logged 46 outfits, cataloged a 222-item inventory, and scored what worked. It went quiet in June. But I think that’s because I’ve taken on those behaviors myself. The tool taught the pattern and then became unnecessary. I need to validate - I’m suspicious of that conclusion - but it’s a case where the kung fu downloaded into me, and didn’t stay in the file.

• Presumed dead. The fitness suite I mentioned earlier. Ten commands for regular use, zero recorded invocations.

...except when I went to look back, even the fitness suite’s logbook told a different story. There are dated entries over several months. Including some that are inside my measurement window. So for some reason the telemetry was blind to it.

The audit result then suggests that there are zero confirmed “dead” skill suites. Silence by itself isn’t enough to declare them dead.

That telemetry issue needs investigating. Measurement brings discipline to judgment. It doesn’t replace it. Apparently a death certificate needs a second witness.

Research - my busiest, most unread, most vital skill

By volume, my deep research skill is the most productive by a lot. It’s made hundreds of artifacts over the last few months.

I almost never reread them.

But that’s the wrong model. If I use a skill to generate a competitive battlecard, its value is clear: someone opens it and uses it.

If it’s a research brief, its value comes from being transformed into something else. After that, it might never be opened again. The value moves forward.

I ran a trace on my ninety research runs. Eighty percent of them led to something real. Thirty-three of them became skills or skill suites themselves. Some of my most-used skills come from the research run that designed them. Others fed identifiable decisions. And some of them led nowhere - like fully drafted skills I never moved to install. No entire suite was confirmed dead. Plenty of individual work was.

The 20% dead rate helps me believe the 80%.

We have to learn to say goodbye

This goes beyond just skills creation.

Build-on-demand works. Twice this summer a real need appeared and I could implement a working Claude skill suite in days, one that immediately became integral.

For personal tooling like this, production isn’t the difficult bit.

But what’s missing - for my practice, but really in every AI-use conversation I see - is the other half of the loop.

Scheduled, honest selection. Kill criteria written at build time, before we get emotionally attached. A quarterly pass where every tool gets a verdict: active, dormant, retired, obsolete - or given a funeral if it’s actually dead.

And I still can’t empirically claim that this makes me faster or better than I’d be if I didn’t have it. I’m not sure anyone can claim that about themselves based on feelings alone. That’s the thirty-nine-percentage-point gap between perception and measured performance in the METR study.

Designing the experiment to measure that is what comes next.

So...I know kung fu. I have the logs to prove it.

But the only kung fu that’s mine is the punch I can still throw after the file closes.

Right now, that’s just one punch. I’ve counted it. Twice.

Further reading:

• Becker, J, et al. We are Changing our Developer Productivity Experiment Design. METR, Feb 2026.

• New data: AI’s impact on engineering velocity is more modest than expected. DX, April 2026.

• Counts, L. AI promised to free up workers’ time. UC Berkeley Haas researchers found the opposite. UC Berkeley Haas, Feb 2026.

Article photo by Getty Images on Unsplash.

Colin smiled at the small boy, who was looking in horror at the cost of a rare gaming card.

“I wish I could.”

“Where are you moving to?”

“Not far. It’s a bit further away from downtown.”

The boy looked past Colin and out the open front of the store. Ruth was opening up the bookstore across the way. The boy clutched a green dinosaur under his arm.

“Is that moving too?”

“The dinosaur?”

“No. The bookstore.”

“Not to the same place. They’re going somewhere different.”

The boy nodded, and put the dinosaur on the counter. “Daaaad!” he yelled.

Behind Colin, boxes covered most of the back wall.

NEW SHOP - FRONT
NEW SHOP - STOCK
STORAGE
DONATE
NEW SHOP - FRAGILE

The last one already had another box stacked on it.

Colin had signed his new lease on Monday. Between a bakery and a pawn shop, just north of Alberta. Two rooms and a dry basement.

Two-year lease, with an option. Not far off three times the rent.

There was only room for one center table. Not room to play.

He’d measured it twice.

More than eight hundred people had paid towards his move.

He could still check the running total on his tablet. It was past target, but there were still new messages.

Thanks for the train set.

My kid still has that rabbit.

Will you still get stuff in from Korea?

Someone had donated a thousand dollars anonymously, just since last night.

The property company’s latest email was next to it.

FINAL OCCUPANCY SCHEDULE - ACTION REQUIRED

He moved it into the same folder as all his other move info.

He watched Ruth drag a cart of books across her sales floor, kicking at a squeaky wheel. The lights were on all the way down to the west wing.

Its entrance was closed off with white construction panels. New ones, printed to look like windows.

A NEW CHAPTER FOR PORTLAND

Someone had drawn a stick figure using the O in Portland for its head.

Colin sold toys. Good ones, generally. From wooden kitchens to dolls that talked. Building sets and board games. Gaming cards and remote-controlled cars. Sometimes used and sometimes new.

No blind boxes.

By noon the record store was pushing music through their shared wall. A pair of teenagers sat on one of the dead escalators, looking through their new purchases from Laney’s stationery shop.

“Morning!” called Dale, walking past in his blue security uniform.

“I think you just missed it,” said Colin.

“Fah. Close enough.”

Dale stopped beside the boxes.

“So when’s the big move?”

“Thursday after close.”

“You need the loading bay?”

“I already spoke to Theo.”

Dale nodded. Kept walking.

The property office had sent up a floor plan with the final schedule. Showed all the occupied units blue, everything else - which was most of the space - beige.

They paid rent to a company in Nevada. Maintenance requests went somewhere. Dale had once had to call the emergency number because water was leaking through the ceiling outside the game shop.

When they’d finally answered, the person on the phone had asked which state he was in.

But they’d also let Colin paint the sign for Good Again himself.

Theo had helped him carry in his glass cabinets the first week, and Ruth helped him get on the patchy wifi. Laney had given him a chair.

It was still there. One leg shorter than the others. A folded catalog under it.

Theo came by with an extra couple of boxes that he set by the register.

“Your new place all signed?”

“Got the keys yesterday.”

Theo tapped the boxes with his boot.

“Laney got a place.”

“She told me.”

“The brothers are taking the game shop virtual only.”

“She told me that too.”

“I think Ruth’s signing something today, too.”

“That’s something. I was worried.”

Theo looked across the corridor.

The café shutters were down. A handwritten sign had been taped over the menu screen.

THANK YOU

“No luck there, though,” he said.

Near the register, Colin was keeping a relocation list on a piece of paper.

Ruth: ?
Laney: Montavilla
Clay House: shared studio ?
Next Turn Games: online / events
Coffee: closing shop

There was a better version on the window over at Ruth’s. Someone had drawn it out on white card. Added the transit stops and listed out how to get from one to the other.

A woman took a picture of it before wandering over to the store. She was carrying a box of magnetic tiles.

“My kids are too old for them.”

Colin counted the shapes.

“Missing a couple of squares, I’m afraid.”

“They might be in the car.”

“I’ll hold the box if you want to check.”

The woman stopped, looked at the moving boxes.

“I thought you’d all be going to the same place.”

“No.”

“Oh.”

The woman looked at the map on her phone.

“Anyone near you?”

“Not really.”

“Pity. That would’ve been nice.”

Colin smiled. “Yeah. I got keys for a new place.”

“My sister says the new area’s nicer.”

Colin stacked the magnetic triangles by color.

“It’s got more natural light.”

“You don’t have windows here.”

He gestured up to the second floor and the ceiling. “Just the ones up there.”

The woman looked toward the corridor. The skylights above the atrium had gone cloudy at the edges. Looked like it was raining outside.

“Bring those squares if you find them. Otherwise I’m happy to put it in the donation pile.”

At the end of the afternoon, the bookstore held its last story hour.

Kids came out wearing paper crowns. Crossed over to the toy store without looking. A couple went to the stationery store.

A girl played with a train set. A boy stuck his nose against the pane of the display case with the rarest game cards. The one from Japan that changed from silver to gold and back every few seconds. Two sisters dressed up a doll.

Their father stood at the entrance drinking coffee.

“Sorry they’re making a mess,” he said.

“It’s no problem. They’re toys.”

The father let his girls bully him into buying the doll. Colin found the price tag.

The father took out his card. The register played the usual holo about connectivity and then accepted the payment.

The girls fought over the bag on the way out.

At six, Dale passed again.

“Five minutes to wrap up.”

Colin nodded toward the children still playing on the escalator.

“Better make it ten.”

Ruth lowered her bookstore gate. Laney pulled her table inside. The music from next door stopped halfway through a song.

Colin moved the chair toward his packing boxes, the catalog slipping out from under its leg. He turned off his own lights and locked up. Went over to take a look at Ruth’s store.

Someone had added another mark to the map stuck to her window. Blue ink.

Outside, the rain was heavier. Colin could see it sliding down the skylights in long, dirty threads.

He lowered his gate.

A story from the Static Drift universe.

Article photo by Albert Stoynov on Unsplash.

But he’d decided it couldn’t just be any old pillow fort. It needed to be different. It needed to be more interesting than “put the cushions against the sofa and throw a blanket over it”.

He wanted me to suggest things.

A cave? No.

A spaceship? No.

An igloo? No.

A pirate ship? No.

And at a certain point, I hit the limits of my imagination. I was tired, and I think he’d just decided to say “no” to everything. He was a challenging client. So I said “why don’t we ask ChatGPT?”.

It gave me ten ideas immediately. Some of them I’d already suggested. But he latched on to the idea of building an animal hospital, and he was off and building.

AI is a good machine for getting unstuck - it doesn’t run out of ideas.

But if there’s a machine that can generate ideas indefinitely, ideas stop being a scarce resource. The scarce resource becomes something else: selection, validation, even taste.

It’s less “can I think of something?”

It’s more “which of these ideas deserves to survive?”

Everyone with access to an LLM has their own brainstorming machine.

Brainstorming fills up a page. It can be energizing. It produces a list of things. But the old refrain of “no bad ideas in a brainstorm” only lasts as long as the brainstorm does. There’s nothing to say that those ideas will turn into something people would use, pay for, or trust.

Idea generation is not invention.

Building a pipeline, not a prompt

I created a set of invention skills that, together, are a structured opportunity discovery and invention suite.

I need the sequence.

Too much AI ideation starts in the middle. Product ideas. New features. Startup concepts. Ten improvements.

We need to start before that.

What’s the opportunity? Who has the pain? What workarounds exist? What would make someone want to use this? What pushes them away from current solutions, or keeps them attached to what they already have?

Once there’s a problem worth exploring, the suite generates concepts that go through multiple methods.

This isn’t free association. This is structured invention.

The suite applies SIT patterns. Uses TRIZ-style contradictions. De Bono provocations. It will collide problems with another domain entirely. It’ll map a solution space morphologically, and build a Zwicky box of parameters and combinations.

And when it’s expanded the problem into potentially hundreds of combinations of ideas, it gets less generous.

Ideas are scored. Stress tested. It compares competing hypotheses against the evidence. Validation planning asks which experiments reduce uncertainty. We do a Mom Test.

We apply kill criteria.

And then the pipeline produces a brief.

One page that states the opportunity, the proposed solution, and the evidence that supports it - provenance is vital. What’s still uncertain? How will we validate? When do we stop?

If you can’t give an elevator pitch for your invention then it’s not finished.

That doesn’t mean it isn’t useful. But it’s not work yet.

Pressure, not abundance

Different methods apply different pressures.

• The opportunity scan will ask if there’s a real problem underneath an imagined solution.

• SIT asks what happens if you remove something, divide it, give it new tasks, or add dependencies.

• Collision takes a domain and forces it into contact with another. Sometimes that second domain might be ancient. Sometimes it’s ordinary. Sometimes it’s weird. The point is not to deliberately get exotic. The point is to apply distance and structure.

• Morphological analysis maps the problem to configurations that might have been entirely overlooked.

• Hypothesis testing weighs ideas against supporting and contradictory evidence.

• Scoring forces those criteria into the open. It attempts to apply some objectivity, and to expose that judgment.

• Validation planning asks what the world needs to show us for this idea to earn more confidence.

• And the final brief ensures that the idea survives to the point of clear explanation.

The suite isn’t magic prompts.

It’s a set of instrumentation, where each instrument marks and deforms the problem in different ways.

Collision rather than metaphor

The skill I have the most fun, for me, is collision.

In a recent piece, I wrote about how old institutions can be repositories of hard-won judgement. Guilds, courts, religious orders, astronomers and scribes. Not because they’re quaint, but because they solved problems around trust, readiness, drift and dissent centuries ago.

My invention suite entrenches that instinct. But it’s not limited to ancient history.

Collision uses anything structurally rich.

DJs building mix decks manage transition, mood and energy. Air traffic controllers sequence risk. Emergency rooms triage scarce attention and resources. I’ve collided ideas and inventions with sources ranging from prehistoric petroglyphs to jazz improvisation to standup comedy.

The source domain has to be useful.

Pure analogy is decoration. “The dashboard is like a city.” Great. Maybe. But something operational needs to flow from that comparison. Metaphors can give the impression of depth without making anything better.

My collision skill tries to avoid that trap by mapping domains structurally.

Who are the roles? What are the processes? What are the constraints? What feedback loops exist? How does the system fail?

After that mapping, it looks for isomorphisms: places where the bones of those structures match up.

I don’t want to know if one idea reminds me of another.

I want to know if one domain has used a mechanism to solve a problem, and whether I can apply that mechanism to a different domain.

Collision should provide functionality.

Making a better dashboard

I’ve been building an internal product flywheel. An AI-and-human pipeline turning customer signals into prioritized work. Scanning sources, routing work, and giving a product manager a draft queue to review.

There’s a dashboard, so that the product manager can see what the flywheel did.

It provided status, trust tracks, run history, signal counts, sizing. It showed what was under the flywheel’s hood.

But the question I was asking of the dashboard was a simple one:

Do I need to do anything?

My dashboard had the problem many dashboards have. It shows you everything you already know. It expects you to infer what matters.

I ran the invention suite against a question: what should the next version of the dashboard look like?

The invention suite came back with a cluster of solutions.

Pattern-driven concepts from SIT. Possible configurations from the morphological pass.

The collider put the dashboard into contact with petroglyphs and twelfth-century administrative registers. That sounds ridiculous - until it produces useful artifacts.

The collision focused on an idea missing from normal critique. Spatial hierarchy is semantic encoding.

If something urgent is visually subordinate, that’s confusing.

That wasn’t the whole answer. But it was an additional pressure brought to bear.

And the same recommendation began to coalesce. A headline-first dashboard, with an action queue separated from the observatory.

That’s a good idea. It might even seem an obvious idea. And the important piece isn’t that the AI produced it. It’s important that multiple methods converged on it.

SIT liked that there was no need to parse the deck first. Morphological analysis had already identified “headline plus detail” as a strong configuration. Collision liked it because spatial hierarchy encodes priority.

It was solving a problem painful enough to justify work. And the hypothesis testing didn’t come back with contradiction.

And so the suite of skills concluded with a brief.

Build a headline-first dashboard with a plain-English status sentence at the top, and in the browser tab title. Put a queue of actions immediately underneath it. Don’t remove all the existing dashboard, but put it below the fold - for observation rather than action.

With a set of validation tests.

Do PMs close the dashboard quickly if it’s all-clear? Do action items get resolved? Does the headline produce false “all-clear” messages?

Scoring, challenging, briefing and validating is invention.

The Machine God isn’t omniscient

This doesn’t make the machine wiser.

The invention suite can still produce nonsense. Scoring can be subjective, with a veneer of precision. Validation plans are useless unless someone runs the experiments.

But the suite provides a foundation for better judgement.

I don’t want any single method to be authoritative. I want it to disagree. I want it to consider a null hypothesis (”don’t do anything”). I want kill criteria.

Curiosity is a good reason to explore something.

The invention suite is intended to help us turn that curiosity into trust.

Back at the pillow fort

My son didn’t need the invention suite for his pillow fort.

He just wanted ten ideas, fast, from a machine that thought faster than a tired dad.

After that, it was just a source of childhood play. The big list of possibilities is great when the cost of choosing is just having to pick up some of the cushions.

That’s not most work.

Product ideas have cost. Engineering work has cost. A strategic choice has cost. AI-generated recommendations entering the workflow have cost.

And as AI becomes increasingly powerful at generating ideas, our judgement of those ideas becomes increasingly important.

AI makes ideation trivial.

That exposes how much of invention was never ideation in the first place.

Invention is about killing the idea you really wanted to like. Then turning the surviving thing into a brief that’s clear enough for someone else to decide what to do next.

Invention is not cheap.

Further reading:

• Why I want my AI projects blessed by Jesuits - on teaching AI what others learned centuries ago.

• Lohr, S. Can A.I. Invent? New York Times, Jun 2023.

• Schultz, B. N. N. Morphological Box (Zwicky Box): Guide with CCA & Example. Service Innovation Labs, Feb 2026.

• Rodriguez, G. R. SIT: Israel’s Answer To Design Thinking? Forbes, Apr 2015.

Article photo by History in HD on Unsplash.

Either the system was wrong, or someone had been operating on Merchant Street for some indeterminate period, and nobody had thought to tell the Council about it.

Pauline had asked for the file. It wasn’t much of one - three pages. The first page was the referral. The second page was a note saying the address had been flagged for inspection five years ago. But that inspection had never been completed due to staffing. The third page was a note from five years before that saying the address had been flagged for inspection. Also never been completed, again due to staffing.

The bus stop was a hundred yards from the premises, and she still managed to walk past it once before she found it. Wedged between a key cutter and a shop that sold - as far as she could tell - exclusively lamps. There was a lovely, green painted sign. A chalkboard below it with the week’s cheeses written in careful chalk capitals.

There were some hanging baskets. Must have been recently watered. They dripped onto the pavement.

She took a photograph of the store front. She didn’t take a photo of the baskets.

Inside, a marble counter ran most of the length of the room. There were three chillers, all correctly labeled. The woman behind the counter pulled out a wooden box at Pauline’s request - temperature logs, handwritten, going back four years. There were allergen labels and batch numbers. The floor was spotlessly clean.

The place smelled of cultured cream and beeswax, and something faintly herbal. She couldn’t place it.

The woman looked to be somewhere in her fifties.

“Mrs. Okoro?”

“You’re here for the inspection?” She turned away from the counter, opening the nearest chiller. “You’ll want to try this hard one. It’s been aging since last year.”

Pauline glanced over her shoulder and wrote down the temperature reading. She accepted the small piece of cheese on a square of wax paper.

The cheese was very good.

It only took twenty minutes to finish her front-of-house inspection. There was nothing actionable.

“Now I’ll need to see the production area,” Pauline said. “And if you could show me your operating documents - licensing, registration, business certification, and so on.”

“Of course.” Mrs. Okoro gestured toward the back of the room.

There was a door. A brushed steel plate on it. Clean, sans-serif, securely screwed on.

Staff Only.

“My grandmother put that up,” Mrs. Okoro said. “I don’t know where she found it. She never said.” She looked at it. “Been there since I was a kid. She put it up and started making cheese. We’ve been here ever since.”

Pauline nodded. Looked at the plate on the door. Looked down at her clipboard. The waiting blank sections for registration numbers and certification dates.

“I’ll need access to the production area, Mrs. Okoro.”

“Yes, of course.” She shrugged. “It’s Staff Only, though.”

“I’m a council inspector.”

Mrs. Okoro smiled.

Pauline held out her laminated ID card. Her photograph, her name, title, and the Council’s crest.

Mrs. Okoro looked it over. “Seems fine,” she said. “If only it weren’t Staff Only.”

“I could issue a formal notice of the inspection. Then you’d be required to provide access.”

“I imagine so, yes.”

Pauline remembered that her supervisor was at a training off-site until Thursday. And formal notices required a countersignature. She moved on.

“Could you explain what constitutes staff, in this context?”

Mrs. Okoro seemed to consider her answer. “People who work here.”

“Is there some way you could designate me as staff? For the inspection.”

“That’s an interesting question.” She shrugged. “There might be a process. I wouldn’t know how to go about it.”

“And the plate itself,” said Pauline. “You’re saying it’s your founding documentation - like a license, in effect. So I’d need to examine it. Could I take it into evidence, as it were?”

Mrs. Okoro looked at the plate. Looked back at Pauline. “It’s attached to the door.”

“Yes.”

There was a pause.

“I’d be delighted to show you everything you’d need,” said Mrs. Okoro. Her tone was warm, almost apologetic. “It’s all right there. Records, equipment, everything.” She gestured again. “It’s just that it’s Staff Only.”

It was the fourth establishment she’d been to that day. The second had had a refrigeration unit that drained, quite inventively, into the parking lot.

The cheese had been very good indeed.

“Would you like something else?” Mrs. Okoro moved back toward the counter. “There’s a soft goat, and this washed rind is seasonal.” She reached for some more wax paper. “I could wrap something up.”

Pauline’s notes reported the compliant shop floor, and the meticulous temperature records. She detailed the appropriate allergen labeling.

Access to the production facility had been unavailable during the site visit. She recommended a follow-up inspection, at the department’s discretion. The earliest slot was twelve weeks from now.

She ate the washed rind on the bus home.

Article photo by Getty Images on Unsplash.

Boring.

Some combination of:

• The Strategic Advisor

• The Risk Analyst

• The Customer Advocate

• The Technical Expert

• The Innovation Lead

A sensible group of faceless people who sound like they’re in a windowless office somewhere talking about operational synergy.

It doesn’t have to be like that.

It could be the Council of Elrond.

It could be a pirate crew.

It could be your younger self, your exhausted present self, your hypothetical 70-year-old self, and someone who thinks your entire definition of success is bonkers.

It could be the contestants of Love Island.

That’s the idea behind my Configurable Council, a Claude Code plugin that lets you run a question through a council of AI perspectives you define yourself.

The process might be serious.

The people don’t have to be.

The council is a mechanism

The original LLM Council, by Andrej Karpathy, asks different language models the same question. They answer, review each other’s answers, and pass everything to a final model for a synthesized verdict.

The diversity comes from the models. Models from OpenAI, Anthropic, Google, and xAI can all make their individual points.

“LLM council” has become a bit of a broader pattern now. Run one model several times, give each instance a different role or perspective. Ask them to approach the problem from different angles.

The diversity is from the prompt, not the model.

The Configurable Council uses that second pattern.

Each member of the council answers separately. Then the answers are redistributed and they review them - anonymously. A chairman weighs the arguments and produces a recommendation.

But you can decide who’s on that council. And that changes more than just the report’s title.

Characters are compact thinking models

A default council might be sensible enough.

A Contrarian tries to find the fatal flaw. A First Principles Thinker asks whether you’re solving the right problem. An Expansionist looks for missed upside. An Outsider catches assumptions. An Executor asks what anyone is actually going to do next.

Professional. Entirely defensible.

But it’s not the Council of Elrond.

Gandalf challenges the framing. Aragorn focuses on practical execution. Boromir looks for power and opportunity. Frodo asks who’ll carry the burden. Galdor demands evidence. Elrond makes the final decision.

The process is the same.

The report is much more dramatic.

But the theme isn’t only decoration.

Any good character is a compact thinking model.

We already understand that Boromir will be drawn to power.

We understand why Frodo cares about the cost to the person doing the work.

We expect Gandalf to tell everyone they have misunderstood the problem.

So we already have a sense of their perspectives.

A “Risk and Governance Advisor” sounds important, but I don’t know how it really thinks.

I have a better idea of how Boromir thinks.

It’s not about more responses

Claude can already generate five answers to the same question.

What we’re trying to do with the council model is generate productive disagreement.

If everyone is kind and helpful, and they produce slightly different versions of the same answer, nothing really happens.

A council needs tension.

Someone should be arguing for the high-risk approach.

Another should think that’s irresponsible and dangerous.

Someone cares about speed. Someone else cares about precedent.

If one person wants to know if the plan will work, another might be asking if success creates a result worth having.

The Configurable Council lets you define explicit tensions. And that disagreement is vital.

A Contrarian isn’t there to add balance. They should be actively trying to kill the idea.

The Expansionist isn’t there to be a cheerleader. They should be trying to push you beyond the caution that undermines the idea.

That’s the work we want.

There is no single council

For a technical architecture question, you might create a council made up of a security engineer, an operator, a maintainer, a finance lead, and the person who gets paged at 3 a.m.

A product strategy question might need a customer, a salesperson, a competitor, a skeptical CFO, and a user who has already decided to leave.

When you’re making a career decision, maybe you need your ambitious younger self, your exhausted present self, your partner, your future self, and someone who thinks you are asking the wrong question entirely.

And some decisions might genuinely benefit from the contestants of Love Island.

Probably not because they know more about corporate strategy.

But expertise might not be what the decision is missing.

Maybe two departments keep saying they’re aligned, but they’re behaving like they’re waiting for someone better to walk into the villa.

Or one team thinks their relationship is an exclusive strategic commitment, and the other behaves like they still want to explore.

Different groups will notice different things.

Your organization probably already has the same groups of people coming together in the same rooms. People with similar backgrounds, incentives, and vocabulary.

Creating a bunch of AI personas that imitate those people doesn’t really give you much diversity.

A dwarf, a hobbit, an elf, a wizard, and a ranger might.

It’s still your verdict

A council is there to provide reasoning.

It’s not there to make the decision.

It’s still AI. A polished report, reviewed responses, a confident-sounding chair.

None of them have to live with the consequences.

Your AI council shouldn’t be an authority. It should expose the assumptions, objections, and missing perspectives around a decision.

It might give you a better answer. It might help you understand why you disagree with the answer it gives. Either will be useful.

The default council is fine.

The Council of Elrond might be better.

Sometimes what your product strategy really needs is a fire-pit conversation with eight people in swimwear trying to work out who’s here for the right reasons.

You don’t always need the same people in the room.

You don’t always need more experts in the room.

You might just need different people in the room.

Further reading:

• My Configurable Council on GitHub.

• Krishnan, R. LLM councils show groupthink. Strange Loop Canon, Jun 2026.

• Champagne, S. What’s Really Behind ‘Love Island USA’ Drama? A Psychologist Explains. Today, Aug 2025.

• Jacobs, Prof A. Why Gandalf and Elrond were wrong. The Homebound Symphony, Aug 2011.

Article photo by Muhammadh Saamy on Unsplash.

Tara had about twenty minutes before the procurement meeting. After that, her afternoon was packed. Urban Continuity at four. Drinks at the Meridian Genomics reception at six.

Twenty minutes was enough to take a quick look at the Walthamstow issue.

It was pretty clean. No heritage claims. Low media sensitivity.

Awaiting review.

Walthamstow Renewal Node 4C

There was an aerial image. Rooflines surrounding a square of tired grass and concrete. Four hulking concrete rectangles. A handful of vehicles parked out front.

Some exterior pictures. Downing Court said the lettering on the wall.

Standard legacy housing stock.

The file didn’t cite the name.

Legacy Public Housing Asset
Maintenance Cost: Adverse. +7.2%/year
Occupancy rate: 39%
Redevelopment Suitability: High

She scrolled further.

Community safety projection: +7.0%
Local spending forecast: +2.8%
Mid-level retail viability: +21.4%
Premium retail viability: +3.2%

Seven percent was good. The model had been cautious since the Lewisham inquiry. Optimistic projections printed beside photos of riot shields and bloody faces had been a company-wide embarrassment.

Two point eight on spending was modest. Might sound more honest.

She checked the residuals in a separate tab.

Households: 22
Median Placement: 13.7 miles
Receiving Authority: Barnet 41%, Ingatestone 59%
Possession Recovery Window: Standard
Appeals Exposure: Low

There were two exceptions flagged. She tapped her screen.

Exception: Mobility-adapted tenancy
Resolution: Assess after relocation

Exception: Caregiving address dependency (unregistered)
Resolution: No tenancy constraint

Simple enough.

It was raining harder outside. A cleaning drone was clinging to the building opposite, battling the downpour as it cleaned the windows.

Tara checked the options.

- Proceed
- Proceed and note additional conditions
- Escalate
- Return to model review

She remembered her quarterly review. Arun had been mostly positive. And kindly.

“You’ve got solid instincts,” he’d said, picking seeds out of his lunch salad. “But you need more confidence in the model. If it’s clear, we should be too.”

Tara had nodded.

It was hard to argue this case wasn’t clear.

She opened the rationale. The suggested text was in pale gray, for review.

Approval supported by projected community safety improvement, local economic uplift, and acceptable relocation burden across resident households.

She changed “relocation burden” to “placement handling”.

Better.

She approved it.

The system brought up the district map.

Node 4C and a thin surrounding catchment turned from orange to green. The color bled outward like dye in water.

The forecast refreshed.

Legacy residential drag: Removed from forecasts
Relocation logistics: Routed

Another notification appeared.

Low-friction approval performance credit applied.
Available in next pay period.

The money wasn’t much. But not nothing.

A good dinner at a good restaurant.

After the procurement meeting, she had a message from Arun.

Good work on 4C. Clean outcome.

Tara typed:

Thanks. Model was clear. Some edge complexity.

She read it again. Deleted “Some edge complexity.”

At six-fifteen, Tara stood in the old magistrates’ court with a glass of white wine in her hand.

The Meridian Genomics account rep had said something about London renewing itself by adapting its inheritance.

Tara nodded politely. She spoke to the account rep about insurance complexities in the genetics industry, and to a former Transport Systems colleague about a new tramline in Docklands.

The wine tasted expensive.

She forgot about Walthamstow. Texted the car service.

The driver took the river route home. They said nothing. The car smelled like leather.

At home, the building concierge recognized Tara before she reached the door. He had the lift waiting.

Tara’s flat was warm. Not too warm.

The kitchen wall had three blinking items to review.

- Grocery substitution approved
- Window cleaning rescheduled due to weather advisory
- Domestic service schedule change request

Fernando wanted to move Thursday evening to Friday morning.

No message. Just the category.

Care Journey Conflict.

The app detailed the impact.

Cost Impact: +£4.20/week
Alternative Providers Available: 3
Confirm Revision / Cancel Service / View Alternatives

Fernando had been with them nearly three years. Cleaning twice a week. Sometimes three if they were hosting. Quiet. Fast.

Also good with Elise’s school projects, which somehow spread across the dining table no matter how often Tara asked Elise to clean up.

She selected Confirm Revision.

A green tick appeared.

Confirmed

The words held on the wall for a couple of seconds.

Then vanished.

Tara opened the fridge. Clean white light. Bottles in rows. Half a lemon in a glass. A child’s yogurt with a cartoon fox on the lid.

Her phone buzzed.

Confirmation of the performance credit. About as much as she’d thought. And the receipt for her portion of the car service.

Tomorrow she had an eight-thirty with Arun about review consistency.

She took a bottle of water from the fridge.

The kitchen wall went dark.

A story from the Static Drift universe.

Article photo by Max van den Oetelaar on Unsplash.

The shape of the work doesn’t change.

It’s faster horses.

I know, because I built one of them.

I made a Claude skill suite called delivery-team. Thirteen role-shaped agents; scrum master, devs, QA, conservative and aggressive project managers - all moving stories through a multi-stage pipeline.

It’s good, too. Fast, tireless, it never lost the thread. It moved “vibe coding” to something with much more rigor around it. It wasn’t something I was looking to replace.

But a couple of months ago I had the chance to sit down with some of the staff at Obvious.ai to talk about their autobuild solution. And an idea got stuck in my head.

It wasn’t “how do I make this faster?”

I wondered if we were looking at things the wrong way. That speeding up human processes is the least interesting things we can do with the power of AI.

Scrum exists because humans get tired, change their minds, need ceremony to keep coordinated. Agents have none of those problems.

So I built a second suite of skills. It’s called flow, and it keeps almost nothing from the first. It trades sprints for convergence. Stories for detailed, executable specs (something Obvious.ai highlighted), pass/fail gates for a Pareto front of competing implementations. No retros, but preserved dissents that reactivate when the conditions come true.

This piece is two things. I’ll talk about the practicalities of each suite - the repo is public, there’s a website explainer, you’re welcome to run either. But it’s also an argument about the underlying thoughts: about the difference between speeding up human workflow or asking what delivery can look like when humans don’t have to coordinate inside it.

Respect to the horse

I built the horse first. And I built it with care.

delivery-team simulates a thirteen person Scrum team, with seven gated stages. They coordinate with documents - agents reading and writing artifacts with clear schema. QA can veto but not write code. Architects lay down high level skeletons that are sharded into stories, and that avoids drift. It uses recognized foundations: BMAD, MetaGPT, Team Topologies.

I can point it at a codebase, or work with it to build from scratch. It goes fast, and it’s good. It never gets tired, never waits for a calendar, and never forgets where it’s at.

But that’s also a problem.

It’s accelerating workarounds that we made for humans. It’s not doing AI-native delivery.

Hell, it still asks me “do you want this to be a 3-day, 5-day, or 10-day sprint?”, even when I know the work will be done in about an hour and a half.

1. Role abstraction is a crutch.

   The suite has thirteen fixed agents. Hand-tuned. But those roles exist because hiring humans is expensive and so we specialize and we commit. That’s payroll, not architecture.
   

   Google and the University of Cambridge’s paper on Multi-Agent Design found that optimizing topology and prompts beats fixed roleplay by nearly 80% on agentic tasks. Anthropic’s own research system dispatches dynamically - one agent for something simple, ten or more for something hard. It’s not using the same org chart every time.

2. Sprints are a backstop for risks in human commitment.

   We time-box sprints because calendar coordination is costly, and if we put things into two-week boxes then it’s safer. An agent doesn’t have any of those uncertainties, and it doesn’t need a calendar.
   

   Sprint boundaries are arbitrary lines we use to manage people. Not the work.

3. Heuristics for humans.

   Story files locked at eighty percent test coverage. That’s a number to trade human effort for human risk. It’s not a property of LLM code. But our stories start drifting the moment we start work.

Rituals aren’t features. And if we put them in the AI loop, they start to become obstacles.

What if nobody gets tired?

flow is designed for agents instead of trying to design around them.

I took some insights from Cognition AI’s 2025 engineering post.

• Actions carry implicit decisions, and conflicting decisions carry bad results.

• Agents can read in parallel safely. But they cannot write in parallel safely, because they all made decisions and accumulate a pile of them that aren’t reconciled.

flow lets multiple agents read, score, search and dissent - but only one of them commits. Intelligence works in parallel, but writing is serial.

Not my invention, but it’s what I built around. It shapes everything that happens downstream of it.

1. Specs are the source of truth.

   I think this is a concept fast getting traction. Take the whenwords library, which shipped in February. More than seven hundred conformance tests, no hand-written code, and all the maintenance is in the spec.
   

   The code is the build artifact of the spec.

2. Stories become generations.
   We build a wide population of implemented variants. They’re scored against an eval suite. It’s not one attempt that we choose to accept or send back.

3. Acceptance gates become a Pareto front.

   We treat quality as a vector. We’re trading speed against simplicity, simplicity against security, and so on.

4. Retros become diffs, reviews aren’t forgotten.

   We maintain a running diff between predictions and productions, rather than the ceremony. And dissents are saved as objects that get reactivated under certain conditions. If a reviewer warns “this will break if we introduce X,” they might be right later.

The bit that still feels like AI being “magic” is what happens next. flow doesn’t build one implementation and refine it. It builds five (or six, or seven - it decides how many it needs) at once, all from the same spec. It scores against different evals, and each build focuses attention on a different metric. Then it converges, keeps what’s good, regenerates and converges again.

That’s five parallel takes being iterated in tandem. If that was a human team, that’s five engineers, five branches, a month of time we probably don’t have. An agent population does it in one afternoon.

It’s a weird way to work. It burns a lot of tokens. But watching it happen convinced me it’s a question worth chasing after.

But we all know the price of gas

None of this is perfect.

flow costs about five times the generation tokens of delivery-team. That signal might be worth every cent. Or you might have spent fifteen dollars when three dollars would have told you what you need.

It fails in different ways, too.

Sometimes it will try to game its own metric. Generators tend to satisfy the eval as much as solve the problem. Goodhart’s Law lives in the same system I built to avoid Goodhart’s Law about old threshholds!

Eval suites can be wrong, and wrong evals aren’t a bug, they’re a problem with the spec. So we might confidently reward the wrong variant.

It’s not fully automated yet (although neither is delivery-team), which is already leaving opportunities for acceleration on the table.

Perhaps most importantly, flow needs that spec, so you need to be able to write it. If it’s doing exploratory work, or you don’t know what “correct” even means, it’s going to flail. And its flailing gets expensive.

The horse is still valid. delivery-team might still be the right tool. When the spec can’t be pinned down. When stakeholders need a recognizable word like “sprint” to trust the machine. You can reach for different skills at different times.

My curiosity has resolved into a much better set of questions. Not yet a verdict.

Curiosity killed the cat

Curiosity is a great reason to build something. But a bad reason to trust it.

So run the same effort through both. You can write the same problem statement. Define the same set of evals. And look at the metrics; time to ship, token spend, human review time, defect count, etc. flow should beat the delivery-team horse, and it certainly shouldn’t regress.

This is still a live test. It isn’t a launch.

Both suites are open-source.

• The repo is at github.com/shinytoyrobots/agentic-delivery-suites.

• Further documentation at shinytoyrobots.github.io/agentic-delivery-suites.

Clone them, run them, apply them to your work. Tell me what’s wrong.

The fork

It’s not really about the specific skill suites.

It’s about the problem. Can you write the spec? Can you instrument the evals? Do stakeholders need a specific vocabulary to instill trust.

One of these is the better horse. The other one might be a car.

Further reading:

• Zhou, H, et al. Multi-Agent Design: Optimizing Agents with Better Prompts and Topologies. Google, Feb 2025.

• Yan, W. Don’t Build Multi-Agents. Cognition, June 2025.

• Breunig, D. A Software Library with No Code. dbreunig.com, Jan 2026.

• Marwala, T. ‘Greatest Good’ Exists Not at the Extremes but Through Exploration of the Middle Ground — Pareto. United Nations University, Feb 2024.

Article Photo by Pete Godfrey on Unsplash.

Emi had arrived early. The owner nodded at her. He set down a hot towel, a small dish of pickled cucumber, and a beer. She was the only customer.

Her phone buzzed. Loud against the counter.

Kenji: Are you watching?

Kenji: Don’t let Hara-san start complaining about globalization again.

The owner read the messages upside down, while he stood polishing glasses.

“He used to be shyer.”

“He was twelve the last time you saw him.”

“That was a good age.”

She smiled.

“You want me to put it on?” the owner asked.

“Yes. He’s already reminded me three times.”

The owner grunted. Pointed a remote at the holo.

Tokyo headquarters. Soft lighting. Dark suits and quiet ties. The stage design was minimal. Expensive.

Kenji stepped into frame holding his cup in both hands.

It was tea these days. Not sake.

The company had stopped using alcohol in official ceremonies after a compliance issue in Singapore a few years ago.

Tea was safer for the international rollout.

Emi stared at the screen. She wondered whether he was still carrying his phone. A wild part of her wanted to text him right now.

His posture was good. Better than last year.

Not perfect. But close enough to surprise her. Somebody had taught him.

Or maybe he’d taught himself.

That might be worse.

Kenji smiled. Bowed his head slightly before speaking.

“Institutions need to care for people,” he said. “So that people don’t need to search for belonging.”

Different offices would be reading different subtitles.

“At Musubi, I don’t think we’re going to replace your family. Your community.” He paused. “But we choose not to reproduce the loneliness that’s so normal in modern work.”

The owner stopped polishing the glass.

“He still really means it,” he muttered.

“Yes, Hara-san. He really does.”

That was the problem.

Twenty-three cities in the stream now. All the new employees gathered in company offices.

They all had their ceremonial cups. White porcelain with the broken circle logo in a deep indigo.

The cups looked cheap. That still annoyed her.

Everyone smiled and bowed to one another. Bowed and raised their cups toward the camera. Hundreds of tiny, not-quite-synchronous gestures from all the rooms around the world.

Family, the subtitles said.

Mutual obligation.

Shared prosperity.

Emi drank her beer.

The people on the screen, as it flipped from office to office, looked embarrassed. Or young. Lonely. Hopeful. The feed from Vancouver briefly zoomed in on a man holding his cup far too tightly with both hands.

She smiled despite herself when the entire room in Edinburgh all answered the pledge loudly enough to distort the microphones. The bartender snorted.

A train rolled overhead.

Three days later Emi was standing in line at her favorite patisserie in Shinjuku.

Kenji: Lunch? A real one, not investor food.

She stopped to read it again, until the woman behind her sighed quietly.

Musubi Tower stood in Shinagawa. Glass and steel and too much chrome, rising above older concrete structures. The lobby smelled of cedar.

There was real cedar somewhere in beneath the AC systems. She remembered Kenji telling her about the consultants who’d come up with the idea.

The receptionist smiled when Emi stepped through the doors.

“Good afternoon, Sato-sama.”

“Good afternoon.”

“I’m afraid he’s running six minutes late.”

“He’s getting better.”

The receptionist laughed.

Employees moved quietly through the lobby. A surprising number of them were carrying the cups.

She saw one tucked beside a tablet computer. Another clipped somehow to a messenger bag. A woman waiting near the elevators was drinking iced coffee from hers while reading a document projected onto her sleeve.

The indigo logo was already half faded.

A young employee held the elevator for her automatically. He was western. Tired-looking. She saw the Edinburgh office lanyard. Visiting headquarters.

“Thank you,” she said as the doors closed.

“You’re welcome.” His Japanese was hesitant.

Recognition flickered across his face a second later.

“You were in Edinburgh last year with Chairman Sato,” he said. “He’d come to talk to the senior class.”

She remembered the trip. Cold and wet. Watching Kenji charm a room full of young faces.

“Yes. The drinks reception in that beautiful old meeting hall.”

The young man blinked.

“That’s right,” he said. “I applied for Musubi’s graduate program the next day.”

“We must be lucky to have you,” she said.

The man managed a thanks before he stumbled out on the eleventh floor.

Emi had to wait until the thirty-second.

Kenji was waiting in the elevator bay with his sleeves rolled unevenly and his phone still in his hand.

“Not so late,” she said.

“Not even six minutes, right?” he grinned.

They hugged.

The executive floor was quiet. It had thick carpet. Lots of dark wood. Semi-opaque glass dividers between the desks.

The company had become expensive in careful ways. She recognized some of them.

An older assistant bowed as they passed.

“Good afternoon, Sato-sama.”

“Afternoon,” Kenji said.

The woman glanced briefly toward Emi as she left.

Kenji noticed.

“You intimidate middle management,” he said, and laughed.

They ate lunch in a small meeting room overlooking the bay. Expensive bentos. Real ceramic plates. No branding on the box.

“You were right, it’s not investor food,” she admitted.

“I try and keep my promises.”

Tokyo shimmered in a summer haze. Lines of heavy freight drones moved slowly toward the waterfront like insects in the sun.

Kenji poured tea for both of them.

Two hands on the pot.

He was still careful.

“You improved your posture,” Emi said.

He smiled. “I wasn’t sure you’d notice.”

“Of course I noticed.”

“I found some old recordings.”

Emi went still.

“What recordings?”

“Some New Year gatherings. Uncle Toshi had a couple of boxes of them in storage.”

Toshi had kept everything. Receipts. Business cards. Tapes and discs in unmarked cases.

“And you watched them?”

“I was curious.”

Of course he was.

“It wasn’t criminal things,” he said. “Just…” He searched briefly with one hand. “Rituals. Structure. People taking each other seriously.”

Emi looked down at her tea.

“Do you think all these other companies out there don’t?”

“They think people are disposable.” He shrugged. “They want people to stay lonely because loneliness gives them more flexibility.”

A quick answer. Not stock. He just believed it.

Below them, a train slid into Aomono-yokochō station, soundless on the other side of the glass.

“You’ll always hate the ceremony, won’t you?” he said.

“I hate the branding.”

“That’s not the same.”

No, she thought. Not exactly.

“All the same things happened after the stream this year,” he said.

She waited.

“Employees connected for hours after the official close. Manila helped Toronto. Helsinki talked to Jakarta about visas. One group in Glasgow got drunk together until four in the morning.”

“With the tea?”

“I’m guessing not.”

She nodded.

“You know we improve retention everywhere Musubi launches it fully,” he continued. “Fewer mental health incidents in three regions last year. People check in on each other.”

“You quantified it,” she said.

“Well, tried to measure it.”

“That’s the same thing.”

Kenji leaned back in his chair.

Outside the window, sunlight flashed across another tower. The reflection enough to whiten their room as it passed.

“You think I built something ugly,” he said quietly.

Emi picked up the cup beside her.

White porcelain. Indigo circle.

Too light.

“Rituals create obligations whether you intend them to or not.”

Kenji frowned.

“But that’s a good thing.”

She almost smiled.

“I’m sure everyone starts off believing that.”

Shareholders were cc’ed on the most important memos.

They called it continuity realignment.

Emi read it on her phone while her driver took her home from lunch in Ginza.

Regional optimization. Operational consolidation.

A small restructuring.

Not from Kenji.

Outside, a Musubi billboard stood over the junction. An old one. The flickering AR overlay couldn’t hide how much the indigo circle had faded.

Her phone buzzed.

Hara: Drink?

The bar was busier than usual. Salarymen hiding from the rain. Two tourists were arguing over the train schedule. A construction worker drinking shōchū. More baseball on the holo.

Hara-san placed a beer in front of her before she sat down.

“Ugly timing,” he said.

She drank.

At the innings break, the holo switched to a business update. Analysts were discussing Musubi’s restructuring beneath a headline praising a “disciplined approach to scaling.”

Kenji’s photograph sat in the corner of the feed.

She didn’t think it was a good picture.

One commentator talked about labor elasticity, and another about shareholder confidence.

“Accountants,” Hara-san muttered.

None of this was scandalous.

The cuts were small. The screen said 1.2% of global workforce.

A support office in Toronto. Translators in Singapore. Temporary workers in Manila moved to zero-hour contracts. Some relocation stipends reduced.

Hara-san wiped down the counter slowly.

“If someone bleeds for you then you owe that debt forever.”

A train shuddered overhead.

“I guess these companies figured out how to skip that part.”

On the holo, an anchor mentioned that Musubi employees had reacted to the announcement with “surprising calm.” They praised the “remarkably resilient workplace culture”.

Hara-san laughed.

Then they cut back to the game.

Emi turned to her phone to find more coverage.

Footage from the London office. A woman leaving holding a white porcelain cup. Toronto, Manila, and Edinburgh.

Small groups lingering after announcements. Huddling. Sharing cigarettes.

Nobody seemed angry.

Unknown Number: Forgive my interruption, Sato-sama. This is Megumi Nakamura. I’m one of your son’s executive assistants.

Emi sat up slowly in bed.

Rain tapped softly against the apartment windows.

Unknown Number: My apologies if I’m asking something inappropriate. Do you still own the Higashi building?

Emi: Yes.

The building was ten minutes walk from Shirokane-Takanawa station. Behind an old grocery, a karaoke bar, and a dog cafe.

It had been years since she’d been here.

It was three stories. Faded tile on the outside. Too many locks.

Too many repairs.

The door stuck.

Emi had brought some groceries. Rice, eggs, beer, instant noodles, and cheap pickles from the station convenience store.

The hallway smelled a little like old cooking oil. Shoes crowded the entranceway.

Voices drifted. Television laughter and people talking. A kettle was boiling.

Normal sounds.

The kitchen lights were on. Cigarette smoke and chatter came from the large living room.

An older woman nearly bumped into her as she came out of the kitchen. “Oh! Sato-sama,” she said, bowing quickly. “I’m sorry to trouble you.” She gave a nervous laugh and stepped aside.

Everything was warm. There were too many chargers plugged into extension cords. She could see shirts drying near the windows.

Someone had taped a cleaning rotation to the refrigerator. There was a list of interviews and train times written in different handwriting.

Most of the people were young.

Japanese. A couple of foreigners. One weary-looking man was trying unsuccessfully to eat curry with chopsticks while two women teased him gently.

The Edinburgh employee from Musubi Tower looked up first.

“Sato-sama.”

“Are you still happy you applied?” she asked.

He shrugged.

“Yes.”

A woman in shirtsleeves stood from the table, and shooed the others to clear some space for the groceries.

“Thank you for letting us use the rooms,” she said quickly.

“Mm.”

Emi remembered when there’d been six men to a room here. Back when they were building the bay expansion. Futons stacked against the walls. Ashtrays everywhere. All the men waiting for instructions that always arrived.

Now the doors to those rooms stood open and she could see suits, laptops, sleeping mats and convenience store bags.

Someone had left a Musubi cup beside the sink. It was filled with chopsticks. The indigo logo was fresh.

“You should sit,” someone said.

“I can’t stay.”

Nobody answered.

From somewhere deeper in the apartment somebody called. “Did anyone get the Wi-Fi password working?”

“Try the old one again.”

“That one was the old one.”

“Try the other old one.”

Someone laughed.

Emi stood near the kitchen while people moved around her.

One young man was practicing interview answers quietly into his phone. Another was helping a woman rewrite her résumé.

The man who couldn’t use chopsticks was explaining visa deadlines to one of the teasing women, using translation software and terrible Japanese.

No one seemed panicked.

The accounting woman lowered her voice and leaned in.

“The company apartments were tied to active contracts,” she said. “A hotel for a few nights is fine, but...”

“But Tokyo is expensive.”

The woman smiled weakly.

“We all know each other from the ceremony groups.”

Someone served out rice from a large cooker. Someone else passed out beer from the fridge.

Emi listened to more conversations about job leads, recruiters, whether it was worth moving back to Sendai.

Nobody talked much about Musubi itself.

One of the younger men gathered empty cans while another woman poured beer into cups from the kitchen.

Most of them were white porcelain. Indigo circles in various stages of fading.

Someone handed Emi one. The beer was cold through the porcelain.

The woman pouring beer paused to raise her own cup just a little.

“To everyone here,” she said, “and everyone still there.”

Emi looked down at the cup in her own hands.

It had stopped raining when she left.

Shirokane-Takanawa station was quiet. A few salarymen drifting home, carrying their furled umbrellas.

Above the ticket gates, a Musubi ad was playing across a curved screen. Young employees, laughing together in a softly-lit office. No sound.

Emi stood on the platform waiting for the train.

On the platform across from her, a young woman in a wrinkled business shirt poured canned coffee into a white porcelain cup.

The indigo circle had almost completely faded away.

Emi’s train arrived.

A story from the Static Drift universe.

Article photo by Nichika Sakurai on Unsplash.

That’s the point. That’s what makes it powerful.

It doesn’t derive its meaning from imperial scale. It’s not filled with golden rooms and marble excess. It has a different idea of power behind it. Borrowed power. Power that’s temporary, held in trust.

The President of the United States lives in the White House, but it’s not theirs. It belongs to the people. All of its occupants are passing through.

Its restraint is intentional. Its plainness matters. Even the name matters.

The White House.

Not a Presidential Palace. Not the Trump House. Not a private estate that’s temporarily on government business. It’s a house that’s white. It’s visible and symbolic, and important in part because it isn’t trying to overwhelm people to prove it’s important.

So the gold matters.

Gold isn’t always ugly. Ornaments aren’t inherently suspect. And I’m not talking about a narrow, museum-approved idea of taste. Neon diners can be tasteful. An eclectic living room full of mixed colors and inherited objects and weird family heirlooms can have taste. Taste doesn’t mean minimal, an absence of excess.

But taste does have a relationship with meaning.

Taste is about what a thing is, where it is, why, what it means, and what it should leave alone.

Trump’s aesthetic doesn’t ask questions. It only asks whether a thing can communicate dominance.

Gold.

Bigger.

Taller.

My name.

More.

Again.

The gold is the easiest part of the pattern to see. It shows up in the Oval Office because of course it does. If something is important, make it shiny. If it matters, make it look expensive. Cover the history with shine.

But the gold, in that setting, does the opposite.

It makes everything look cheaper.

Not because gold is cheap. But because it’s unnecessary. A gold White House doesn’t look richer. It looks less sure of itself. It suggests that the inhabitants don’t trust the building to hold its own value. The Oval Office not trusted to carry its own authority. History that can’t speak without extra decoration.

It’s trying to prove itself because belief has failed.

The spectacle isn’t meaning. It’s a substitute for that meaning.

You don’t need to say you’re cool if you’re cool. You don’t need to tell people how important a room is if it’s already important. You don’t need to plate things in gold to give them meaning if the meaning is already there.

The gold makes the real things look fake.

And we’re not just talking about gold. It’s a bigger ballroom. A bigger arch. Triumphal structures that must, in cost or scale, outdo other triumphal structures. Monuments can’t stand, they have to dominate. It’s the belief that you make things more important by making them bigger.

Greatness is not volume.

We don’t measure how good something is by its square footage. Not the height of an arch or the size of a room or the amount of reflective surfaces you’ve added to it.

Greatness not only can be quiet, it sometimes has to be.

The more a leader needs to tell everyone how big and great things are, the smaller the leader looks. The more a government needs to tell people how confident it is, the less confidence it seems to have. The more a country has to keep saying how important and strong it is, the more fragile the performance.

It’s almost pitiful, because it can’t ever stop. A room can’t be a room, a building can’t be a building, and a country can’t be a country. Everything is just a setting for a repeat performance: look how important this is! Which really means, look how important I am.

That’s not confidence.

Bigness is insecurity with a huge construction budget.

Donald Trump’s been called a poor person’s idea of a rich person. I get the idea, but it’s not quite right. It’s not about poverty, it’s about fantasy. Donald Trump is insecurity’s idea of a rich person.

Gold things are rich. Tall things are important. Putting my name on things means they’re owned. A big room is powerful. I need people to see it from far away so that it matters. If my name is on it then nobody will forget who made it.

This isn’t aspiration. Aspiration can be many things. Playful, hopeful, silly. We can aspire to comfort, beauty, abundance, spectacle, shine.

This reduces value to what can be outwardly signaled.

And if it’s private, it’s pretty harmless. If someone wants to live inside a gold apartment, all power to them. If someone wants their bathroom to be made just of mirrors and marble, great. Everyone should be able to inhabit their fantasies.

Public space is different.

The White House isn’t someone’s private fantasy. Washington, D.C. shouldn’t be a personal mood board. National monuments aren’t brand extensions. The republic isn’t a developer portfolio that needs nameplates.

A gold apartment is a preference. A gold republic is a warning.

It’s not a neutral aesthetic. It speaks to ownership. It speaks to stewardship. It says what the person holding power for now thinks that power is for.

Gold says that this is valuable because it shines.

Scale says this is important because it’s big.

My name on a building says that public memory isn’t enough.

Demolition says that I’m more important than what came before.

And we might, just, take each of these in isolation as vanity. But together they pull together into a worldview.

Possession.

Not care, or continuity. No humility. Ownership.

If this thing exists then it can be claimed. I can redecorate this room into submission. I can rename this. I can overwrite history into something larger, shinier, newer, and more useful to me personally.

But that’s not how inherited meaning works.

The Oval Office, the White House, the presidency, they all matter without Donald Trump. America - as symbol, story, argument, and promise - matters without Donald Trump.

Inherited meaning gives you a role but it doesn’t make you the author of the thing.

Trump’s aesthetic wants to reverse that. He doesn’t want to inherit meaning. He wants meaning to inherit him.

But the people’s house shouldn’t point toward one man.

That’s so small.

That’s the smallness inside the bigness.

It claims grandeur. But selfishness is cramped. Everything pulled toward the person until there isn’t room for anything else. No room for history. No room for the public. No room for any idea that something might matter more because it doesn’t belong to you.

That’s tacky.

Not the decoration, the shine, or the theatrics.

Tacky is decoration that doesn’t know what anything means.

And that ignorance is dangerous in the public setting. A nation depends on the idea that it has things held in common. Some borrowed. Some inherited, with inherited obligation. Things that aren’t raw material for self-expression for whoever’s holding the keys right now.

You don’t gild a house you don’t own unless you’ve forgotten whose house it is.

There is a broader argument here, about patriotism and fragility, and the fear of a country changing. This is the surface where it’s most visible.

The same insecurity that needs gold also needs walls. The same fear that wants personal monuments is also scared of history.

Confident countries don’t need this.

A confident country doesn’t announce its importance all the time, because the importance is already there. The institutions, the history, the contradictions, the people, and the unfinished argument of what the country is.

Confidence is happy to leave room.

Insecurity will try to fill every surface.

Gold isn’t always wealth. Sometimes it’s fear pretending to shine.

Further reading:

• Post, J. & Doucette, S. How Donald Trump’s Narcissism Masks His Extreme Insecurity. Literary Hub, Nov 2019.

• Maddinson, T. Colonial Statues. Public Space and Masculinity in Postcolonial Britain. Cast in Stone, Dec 2012.

Article photo by MUILLU on Unsplash.

His tablet didn’t take his face the first time. The light too dim. He had to angle his face in the light of the strip above the shack door before the circle went green.

Asset protection shift: active.

It was chilly. Not freezing, but he felt it in his knees. The zipper on his jacket caught on the torn cloth beside his company patch. He worked it loose with his thumb before he stepped out into the yard.

Cars on both sides. Three high. Four in some places. Sedans, work vans, fleet pods with their windows blacked out. There were delivery carts with subscription codes still attached. Useless without an account.

A reclamation site, they called it.

Nothing got reclaimed.

The fog swirled gently, caught between the stacks. The morning dew caught on the hoods and made lines in the dust. The cars all had tags, on the windshield or the door. Blue, yellow, and red. Processed, pending, and restricted handling. Half the tags had faded to white in the sun anyway.

Hank started up his buggy.

Checkpoint A1. Green.

Checkpoint A2. Green.

A crow perched on a rusted minivan cocked its head as he went by.

The east side fence was patched. Mesh over mesh. Nobody had interfered with it since he’d reported the damage last week.

Unauthorized material access attempt. Increase visible patrol.

Hank walked to the fence and prodded the mesh with his toe.

The gravel road went past the compacts. He looked over them at the old parts plant. Behind its own chain link fences. You could still see the ghost letters on the wall where they’d taken the company name down on his last day.

He slowed the buggy at C7.

One of the tarps had blown back. Not much. But there was a flash of color. A deep red that peeked through the dust. Not flat corporate warnings. Something bright.

He walked over, pulled the tarp back.

The plastic scraped over the rust.

A long hood and a wide dash. The mottled chrome around the grille caught in the dawn light. The windshield was cracked but not gone. The seats inside were split, dirty yellow foam pushing out against the vinyl.

Hank put one hand on the fender. Let it rest there.

Cold metal. Weighty under his hand.

He looked back down the aisle. Back toward the camera post at C5. Its status light was blinking amber. It’d been like that for months, but service ops insisted it was operational.

He pulled on the door handle.

Heavy. A screech of complaint from the hinge before it gave. A clunk as it opened.

Mouse nest. Old vinyl. Mildewed foam. Oil.

Hank leaned in, one hand on the roof, and looked at the dash. Gauges. No screen. All the needles resting behind cloudy glass.

There was a key in the ignition.

He laughed. “Course there is.”

He went around to the front of the car and popped the hood. The latch fought back, and the hood rose heavy until it settled on its prop.

The engine was still there. Usually they got stripped down to the bolts. This one sat. Big block. Belts you could see the cracks in. No battery, and a bunch of chewed insulation.

He’d made steering-column brackets in the plant. Used to be for cars like this. Then for the small corporate carts with the round nose and sealed panels. They stopped if the monthly authorization failed.

They were still good brackets.

He had a jump pack in the patrol buggy. “Emergency site mobility support,” said the label. The cart died twice a week and corporate wouldn’t replace the battery.

He carried it over to the car. Set the pack on the frame and cleaned the cable ends with his pocketknife. Not well, but enough.

He fetched the little fuel can from the buggy, too. A capful, maybe even less.

His hands knew the order.

The tablet vibrated.

Warning: PATROL INTERVAL EXCEEDED.

The driver’s seat dropped under him when he sat, with a deep sigh of dust. He put a boot on the brake. Muscle memory.

The key turned. A click.

He stopped. Turned it again.

The starter dragged like it was pulling itself out of the mud. One slow grind, and then nothing.

He waited. Listened to the wind.

Turned the key a third time.

It wouldn’t be accurate to say the engine started. But it moved. Two turns, maybe three. A hard cough coming up through the block, deep enough to hit the floorboards. The car shook under him. The dust lifted from the dash. The needles on the gauges shifted.

It caught. It barked once.

Then it died.

The echo rolled back down the aisle to him. A smaller memory.

Hank sat with the key in his hand. He rubbed the steering wheel with his thumb.

The tablet vibrated.

Audio anomaly detected. C-sector.
Confirm incident status.

He got out slowly.

He unclamped the jump pack and wound its cables. Put the fuel can back in the buggy. He lowered the hood, gently, and let it drop the last few inches.

It latched.

The tablet buzzed again, options waiting.

Asset tampering.
Equipment malfunction.
Unauthorized access observed.
No incident.

Hank looked down the aisle. The red car sat beside him. Dull again, shrinking into the torn tarp. The gray carts watched from their own stacks.

No incident.

The tablet accepted it.

Checkpoint C-8 blinked in its map.

Hank pulled the tarp back over the car as best he could.

A story from the Static Drift universe.

Article photo by Tim Mossholder on Unsplash.

Earlier this year this centuries old approach to astronomy and astrology solved a problem I had trying to fix a dashboard.

The dashboard’s mine. It’s on top of a system I’ll come back to later - an AI pipeline to turn customer signals into executable work. The dashboard tells me if I can still trust the machine’s judgement.

It had a problem every dashboard faces. It can’t see its own drift.

When the AI slowly gets worse, and if my own instinct on “good enough” also slides, the dashboard will have green lights all the way down. It’s not lying, but it’s quietly going wrong.

The astronomers at the Jantar Mantar solved the problem centuries ago.

They knew their instruments drifted. They knew the assumptions they had in their calendar would, over time, pull away from the actual stars in the heavens.

They also didn’t trust the running system to catch its own decay. They re-anchored, on a schedule, against a fixed external baseline. It’s a mathematical correction called Ayanamsha to anchor their astrology to the fixed stars, and account for the earth’s drift.

I didn’t learn this solution from a digital product blog about dashboards.

But I did get it on purpose. By colliding the problem in front of me with a domain that had nothing to do with it. And, once I really started doing that deliberately, I can’t stop noticing a pattern. The hardest problems in the newest technology - AI - are often old problems wearing new clothes. And ancient answers are better than ones we’re busy reinventing.

The library five years shallow

Most people building with AI are reasoning from five years of data at most. Last quarter’s framework, the most recent SaaS playbook, or the pattern that worked last time.

That’s not a knock. It’s a fast moving field. Five years ago feels like forever. But it means we’re solving ancient problems with a shallow frame of reference.

And it’s not just a story about code. That’s the one we’re telling the most. AI writes a function. AI reviews a pull request. AI ships a feature. That’s real, interesting, and only a slice of the whole.

I lead product. When I’m using AI it’s not primarily about writing code. I’m triaging customer signals, internal data, and predictions about the future. And using that data to recommend how to route, draft specs.

How to decide what’s worth building.

Those are judgement calls, and I need to know how much of that judgement I feel comfortable trusting.

Is something ready? Calibrated? True? Is that prediction honest?

These aren’t engineering problems. They’re the problems guilds, courts, councils, and churches have stress-tested over centuries - and we can use the answers they wrote down.

So I went looking for the answers on purpose. This is how I did that, and what I found when I started building with them.

Reading old code

The how is a method I built into a skill called inv-collide. Part of a broader suite of invention related skills.

It stems from an idea called bisociation, created by the author Arthur Koestler. Then it operationalizes that idea.

Take a problem in front of you, and then a domain that has nothing to do with it, and you force them through three steps.

1. Map both as structures. The roles, processes, constraints, feedback loops, value flows, and failure modes. Not what they’re about, but about the bones of what they do.

2. Find the places where the bones are identical. Isomorphisms.

3. Generate concepts at that intersection. e.g. if this domain solves problem P with mechanism M, and I also have a version of problem P, can I transplant M?

It’s not the same as brainstorming. Brainstorming is free-association. Bisociation matches the bones of the thing. That discipline makes the output usable.

And that discipline requires a step people might skip.

Disciplined enough to discard it

It’s all very romantic. I found a poetic parallel between a dashboard and the stars in the sky. Well, the universe is really big and it’s easy to make up a metaphor.

If the method was just making nice-sounding coincidences, it would be a party trick.

It also generates a big discard pile.

When I collided the trust dashboard with the Jyotish astronomy and the Jaipur observatory, it surfaced twelve structural matches.

Astrological muhurta - an auspicious window where conditions align and you can move forward - mapped directly onto a promotion gate in my system. The point where AI capability became more trusted to act autonomously.

That got thrown out. It wasn’t bad, but my system already had a streak counter and encoded readiness gates. This added costume, not structure.

The collisions are only worth integrating when they survive an honest attempt to kill them. Re-anchoring survived. It runs in production.

Running old code in my product flywheel

Underneath the status dashboard is what I call my product flywheel. It’s an AI-and-human pipeline to turn customer signal and internal consensus into execution ready work, without a person writing every issue.

It reads from various sources - customer knowledge base, Slack, Zendesk. Then it classifies and routes what’s changed. It’ll draft a business case and independently assess it, scaffolding the approved work into Linear. Then it will recommend routing, priority, and provides a rate-limited queue for engineering to pull from.

This is product work. The product flywheel feeds execution, but it isn’t the execution. This is triage, routing, specs, and prioritization. Engineering pulls from its output in order to execute.

This isn’t a story about code review. These are rules for intelligence from medieval guilds, eighteenth-century astronomers, and the Talmud. And they’re applied to the most difficult parts of trusting AI with product judgement. Applying methods that were created a long time ago.

Here are three that are running.

Earned trust, and the medieval masterwork

One of the oldest problems in management. When do you let someone work unsupervised?

If you get it wrong early, you’re letting unqualified hands do a lot of damage in your name.

If you get it wrong late, you’re throttling the potential of someone who was ready.

Every craft tradition that’s lasted seems to have solved this in the same way. With a gate. A guild apprentice submitted a masterwork, and the sitting masters judged it. A Jesuit student reached a point where his judgement was, in their words, formed. Trust was domain-specific, slow to build, easy to lose.

AI tooling uses a flag. We set a permission mode, and we crank the autonomy up and down depending on how brave we’re feeling. But the system isn’t demonstrating anything.

The flywheel doesn’t have a dial. Its agents earn the right to act, and they don’t start with it.

I track trust per stream. Incoming customer signals, routing, drafting, the queue controller all have their own standing. Being good at one thing is no evidence of being good at anything else.

Each capability can climb through three tiers. The tier changes what the agent is allowed to do. An apprentice intake agent reports what it would have done, line by line, and needs a human to confirm every call. A journeyman provides a summary of what it would do, and asks for a single confirmation of that summary. A master executes and then reports back for review.

An agent moves from apprentice to journeyman on fourteen consecutive clean runs - confirmed, by a human, as correct. The journeyman to master takes thirty. And this isn’t an average. One bad run resets the streak to zero. And mistakes made by more “senior” agents mean demotion and a doubled threshold.

That’s the medieval guild structure, as a YAML file.

Trust is isolated by domain. It’s earned slowly, lost easily, and more expensive to win back a second time. That’s not how permission flags work, but every master craftsman who ever lived would recognize it.

That makes autonomy something earned - and easily lost.

Calibrating my judgement against the stars

Let me close the loop on that dashboard.

I’m not worried about the AI “breaking”. That will be loud and obvious. What I’m worried about is silent drift. Miscalibration as the AI degrades slowly, and the gap never shows up.

Those astronomers in Jaipur had an answer. Re-anchor on a fixed, external, baseline. On a schedule. Not trusting the instrument to audit itself.

My flywheel uses two versions of that.

1. A deliberately imperfect target. My runs are clean if I override the AI’s call no more than fifteen percent of the time. Not zero. Zero overrides is a person not paying attention. We want a target that keeps someone in the loop.

2. You can’t re-baseline a scoring system - even if it’s an improvement - without re-scoring at least five historical projects against the new rules. Then recording the sign-off and noting the discontinuity. You can’t move a baseline and also erase the evidence that you moved it.

That’s Jantar Mantar. Re-anchoring against a fixed point, on a cadence. And keeping its receipts.

Adversarial truth-seeking, in the room and in the spec

If we review for consensus we throw information away.

Two smart, competent people disagree about a tough problem. When that disagreement gets resolved, we move on.

But that signal tells you a problem has more than one shape. And that losing argument might be right under conditions that develop in future.

The Jesuits formed students through disputatio. A structured, adversarial defense. You don’t prove you’re competent. You prove you’re competent when a skeptic attacks your work.

The Talmud has preserved the minority ruling alongside the majority one, deliberately, for two thousand years. A defeated argument might become the right argument when the world changes.

Those memories need to be written down. If they don’t, we don’t remember them. A dissent gets aired in a meeting, a decision gets made, and the losing argument is - at best - noted in a retro document that nobody reads. That’s even worse for product decisions - routing, prioritizing, build this not that - than code, because the record’s thinner.

The flywheel runs these institutions against product judgement.

Disputatio is in every routing decision. Recommendations don’t come from one agent, they come from a pair. One proposes the route and the priority, and the second is designed specifically to challenge it. Attack the recommendation before it’s committed. The proposal has to survive the examination. That’s Jesuit insight - that the defense is where the judgement is formed - applied to a decision about a customer request.

Chavruta is the basis for how the system records the disagreements. Reviews produce dissents. The dissents don’t need to be resolved, but they are committed. And not just as a stale record. They’re structured objects that include conditions under which they wake up again. Recheck at the next live run. Resurface if a specific metric flatlines. When the world changes to match a trigger, that old losing argument comes back on its own and demands a second hearing.

Consensus is lossy. The Talmud knew that two thousand years ago. My flywheel acts on that memory.

Others still on the bench

There’s a couple more collisions which produced things that I haven’t shipped yet. The same method, but in the design stage.

1. Drift lineage. A text copied by hand for a thousand years has scribal drift - small errors that harden into official fact because the chain is trusted. Traditions that survive have apparatus for this - lay one manuscript next to another and see where the text mutated. I’m building the same thing for a claim - a walk backwards to the primary source, and flags where the numbers or the conclusions changed. Not whether those changes were right or wrong, but a lineage of provenance and drift from source.

2. Avoiding “vaticinium ex eventu“. This is prophecy after the fact. Re-reading the record once you know the outcome, and bending the record to show you were right. AI multiplies confident predictions, and decision journals are editable. Medieval clerks had the answer. A boring, comprehensive one. Witnessed, dated, tamper-evident entries. Predictions logged and stamped, and scored at outcome time. It’s about not lying to yourself about a prediction.

We keep reinventing what was proven

It’s about trust.

Transferring trust to something that acts in your name. Keeping your own judgement from drift. Not letting disagreements die. And being honest about predictions.

None of them are about AI writing code.

It’s about trusting AI’s judgement, calibrating it against my own, auditing it, and keeping us both honest.

This match to old institutions is real, not a flourish. History is not just a charming source of cool metaphors. It’s where they already ran these experiments. Where they wrote down the answer in a language nobody speaks any more.

It’s the debugging history of our species, and we’re reaching past it for last quarter’s new framework.

When you hand an AI something that matters - a decision, a forecast, a release - find the institutions that already solved it.

Read their old code.

Get the thing blessed by Jesuits.

Further reading:

• Sarda, S. India’s mysterious gateway to the stars. BBC, May 2022.

• Popova, P. How Creativity in Humor, Art, and Science Works: Arthur Koestler’s Theory of Bisociation. The Marginalian, May 2013.

• Farrell, A. P. The Jesuit Ratio Studiorum of 1599. Conference of Major Superiors of Jesuits, 1970.

Article photo by George Dagerotip on Unsplash.

I texted back.

finally

He replied.

right?

That was the whole conversation.

I wanted to go over. But I thought he should have time with it. He’s waited long enough. We both have.

When Sunday comes, I let myself in like usual.

“Leo-?”

He’s at the kitchen counter. He looks up and grins and I see it. Gray straps across his chest and right shoulder, some kind of darker webbing. A small white module that sits just below his neck.

There’s a blue light pulsing slowly.

I think it looks like someone made a wearable commercial fire alarm.

“Look at you,” I say.

“I know.” He spreads his arms. “Two years and eleven months.”

He’s so excited. I’m holding my breath. Maybe he can manage a week alone. I don’t know what to do with that.

“So are you going to show me?” I say.

He takes me through it. His CareFrame.

The sensor is under the skin of his forearm. A silver disc, smaller than a watch battery. That’s the Morizono bit, so he can show off and let the filigree catch the light. He had it...installed...on Thursday. The skin’s still red.

The harness is Synetica. Less glamorous, but it’ll probably work forever. The module tracks biomarkers, logs his meals through a relay. It manages his medication schedule, tracks his energy registers and tells him when to rest.

“It’s like having a nurse,” he says. “But one who doesn’t sleep or get paid.”

I smile. “Oh, they’re still getting paid.”

He grins. It’s the most excited I’ve seen him in a while.

I busy myself making lunch. The pasta with the broccoli rabe that I know he likes. Lemon and breadcrumbs. I remembered to pick them up from the shop on the way. I’ve been bringing him his groceries for a decade.

Maybe the CareFrame will help him remember to buy ricotta.

I put the plates on the table and he sits down. His harness chimes. Not loud. A single note. Not much more than a regular phone notification.

He glances down at his device. “Says I should have more protein.”

I look at the plate. “I put chicken in it.”

“Not skimping me, are you sis?” he says.

He eats everything.

He checks his module afterward. Logs the meal. The blue LED pulses twice. Apparently that means it’s satisfied. I watch him do it. I wonder if it’s read all the books on nutrition that I did.

After lunch he puts the kettle on. The CareFrame chimes again. This is a different pitch, longer. A bit more discordant. He looks at the module, then at me.

“It’s a rest reminder,” he says.

“Okay.”

“It’s probably fine. I can ignore it.”

“It’s supposed to take care of you,” I say.

He sits down on the sofa and the module pulses in approval. I finish making the tea. I sit across from him and there is nothing else for me to do.

“So...”

I’m at a loss. It’s my job to notice. All we ever talk about are early signs. Creeping weight loss. Whether he looks tired or not. I can read it in his voice. It’s been a decade.

Is that all we have to talk about?

The CareFrame doesn’t need to hear his voice. It just knows.

I sip my tea.

“You’re quiet,” he says.

“Just tired.”

“That’s my line.”

I suppose I can go to work tomorrow and not worry. Might be able to have a date that doesn’t slowly falter because I have to watch my phone. I stopped dating to be Leo’s primary contact.

Is that me? And is CareFrame better at it?

Leo doesn’t push.

We half-watch a movie, neither of us really saying anything. It’s starting to get dark outside. The CareFrame’s chimed three more times. Medication, hydration, and a second rest reminder even though he’s already sitting down.

I don’t say anything about it.

When it’s time to leave, he stays on the sofa.

“Will you remember the chili in the freezer this week?” I ask him.

“I promise.”

“And your infusion?”

He taps his harness. “I think we’ve got it.”

I’m at the door when he says my name.

“Maya.”

I turn. He’s standing up now. The harness is creasing up against his shirt. The blue light is steady. He looks at me.

“It’s weird, isn’t it?” He says. “Getting it, finally.”

“It’s good,” I say. “We both needed it.”

“It’s what I needed,” he nods. “But it’s not the only thing I needed.”

He takes a step closer.

The module chimes. A medication reminder. He doesn’t look down.

“I know,” he says. To the module, I think.

He reaches out and his hand is on my shoulder. He squeezes once, and draws me in for a hug. The harness is scratchy and the module bumps up against me. but I don’t mind.

“It never built a treehouse with me,” he says.

The module chimes a third time. He ignores it.

“You’re such an idiot,” I say.

We stand there. The CareFrame chimes again.

We hug some more.

Another chime.

“Yeah, yeah,” he says.

A story from the Static Drift universe.

Article photo by علیرضا افتخاری on Unsplash.

It also seems to have a problem: a strategic imagination still bound to the canvas.

I realize that’s a challenging thing to say about perhaps the most important product tool of the past decade. This is not a “Figma is dead” article.

Figma changed how digital product teams work. It made design a genuinely multiplayer activity. It made a design file a shared space. Collaboration, critique, exploration, and handoff in a browser-based canvas everyone could see.

Sketch looked comfortable before Figma came along. Users and workflows and plugins, and enterprise legitimacy. A whole ecosystem. InVision for prototypes, Zeplin to support handoff. Abstract for version control.

Then Figma came in like the Kool-Aid Man and made Sketch look obsolete almost overnight.

It wasn’t anything to do with Sketch’s design features. It could still draw a rectangle!

But Figma changed the whole basis of where the two products were competing. Not the design tool with the best interface, but making design collaborative.

I’ve never seen another product that created as much practitioner pressure for change as the internal demand at IBM to switch from Sketch to Figma. It overcame corporate inertia faster than I’d have imagined.

Figma just had better answers. Staying on Sketch meant being left behind.

Figma solved the coordination problem of its moment. Its risk is in continuing to solve the problem after the problem has changed.

There is a historical parallel. But it’s not as glib as “Figma is the new Sketch”. That’s too neat. Figma is clearly larger, more deeply embedded, and has a degree of strategic awareness.

But incumbents don’t usually look like they’re sleeping. Especially from the inside.

Figma is shipping a lot of stuff. And they’re telling a coherent story about the future that runs through them.

Are they building that future, or just extending the conditions that made them dominant before?

The center of gravity is moving from canvas to code.

That means from abstraction to execution. From static artifacts to live systems. And from design files to context that AI interprets and generates from.

Designers will still need visual tools. And teams will need shared spaces for critique and exploration.

But what does Figma do when the canvas is not the center of gravity?

How Figma won in the first place

Figma’s first great achievement was technical. They made the browser matter far more for design than anyone thought possible. Cross-platform access mattered. Performance mattered. Multiplayer mattered.

The product was excellent, and execution counts.

But the deeper shift was cultural.

Before Figma, collaboration was fragmented. It needed local files, redlines, PDFs, and those meetings where everyone asked “is this the right version?” Figma collapsed all that distance.

Figma wasn’t merely a better canvas. Figma was a better coordination model.

It made work around the design abstraction collaborative. Which was a huge step forward.

But an abstraction is still an abstraction.

The canvas is not the product. It’s a representation. The real product is in code.

The canvas was vital for helping us think before the reality of implementation got too expensive.

But it depends on a world where there’s a big gap between visual intent and working software. That’s where the abstraction lives.

AI is collapsing that distance.

The canvas answers a translation problem

The canvas makes sense.

Designers express intent. Engineers translate the intent into code. Product managers mediate priority and scope.

We use the thing we imagined to help us ship the thing that’s real.

And that model isn’t going to be going away any time soon. Many organizations will likely work this way for years to come, if they can get away with it.

But the direction of travel has changed.

Design-to-code is faster. Which is great. But it’s just collapsing the way we already work. Handoff, but faster. Translation, but faster.

What’s genuinely different is how structured design and product context, component code, and rules can be interpreted directly into coded, working interfaces. A prompt no longer has to start from nothing if it has access to the design system, APIs, patterns and engineering constraints.

And design becomes that context. A context for AI execution systems to use.

Teams are still going to need visual comparison and critique. They’ll need shared spaces to make business calls. The terminal window or an IDE is not a place for a lot of stakeholders to participate.

That doesn’t make the canvas central.

Bring it back to the canvas

When I look at Figma’s recent moves, they make sense. They build on its current strength.

More work should happen in Figma. More artifacts should come from Figma. Workflows should come back into Figma. More of the product development should be in the Figma ecosystem.

Reduced to its simplest form, the strategy seems to be:

Bring everything back to the canvas. Our canvas.

But the next era won’t be organized around that.

It’s why I thought “code-to-canvas” was pretty revealing. Make a real thing, then bring it back into Figma as editable frames.

That might solve a short-term collaboration problem. Directionally, it’s strange. Actually, it’s wrong. Wrong for the future, even if useful for Figma’s current position.

In that example, Figma is more worried about getting you back into their room - where they know how collaboration works. Less worried about whether that’s the right model of collaboration for the future.

The canvas won’t be the source of truth

Of course, Figma might be moving towards a more compelling future. One where Figma is a collaborative interface that reflects reality.

But it would be Figma as a lens.

Figma might be where you inspect your working systems. Compare variants. Annotate things that are real. See the design system drift. To steer and govern.

That might be valuable.

It also means accepting the canvas isn’t the center any more. And if it remains important, it only does so if it can be an interface to the truth.

The code, the runtime, what’s real, and what actually ships.

Figma’s danger seems to be trying to remain central by making everything pass through your old model.

That’s an incumbent trap.

That’s looking at what made you dominant in the first place, and only working to improve that thing. And that will be right...right up to the moment that the basis of competition changes.

Figma won against Sketch because it realized the center of gravity could change.

Now that center of gravity is changing again. And Figma is on the other side of the innovator’s dilemma.

Execution is cheap. Coordination is not.

AI makes execution cheaper.

Not free. But from a practitioner perspective, it can feel that way.

AI scaffolds the screens, uses the components, wires them up, refactors and gives us variants. It can create at a speed that changes all the old bottlenecks.

So the limiting factor is not “can we produce an interface?”. The limiting factor is “can you produce the right interface, with the right standards, for the right users, in a way our organization can trust?”

Coordination with AI assistance is not the same as collaboration in a canvas abstraction. We need structured and ranked context.

Which components are approved? Which patterns are deprecated? Which implementation is authoritative when the docs say one thing, the code says another, and Figma says a third? Which accessibility rules apply? Which regulatory constraints matter? Which engineering standards are non-negotiable?

That isn’t a canvas problem.

It’s an infrastructure problem.

Design systems are even more important in this world. Not as component libraries or asset stores, or even as docs for people to manually consult. They’re executable intelligence that tell AI systems how an organization builds.

The canvas is insufficient. It can arrange. It can invite critique. But unless it’s deeply connected to some control layer of product delivery, it risks becoming a pretty picture while the real thing lives elsewhere.

That’s a strategic problem.

What Figma seems to believe

From the outside, Figma seems to believe it can expand its canvas to contain the next era.

And, look, that may be unfair. It’s an external read of a company’s strategy. Figma is full of smart people, with every incentive to understand the shift. It may even be the smart commercial decision. That doesn’t make it the right product model for the next era of work.

Product strategy reveals posture. And Figma’s posture seems focused on a return to canvas.

Bring your generated work back. Bring your coded artifacts back. Bring your developers into Figma. Bring AI into the canvas.

Put more of your organization into the place Figma owns.

Which isn’t necessarily stupid. Enterprises have historically liked consolidation. People are familiar with Figma. And Figma has a gravitational pull from its market dominance.

Figma can keep adding useful capabilities.

Will those capabilities help Figma adapt to a world where the working artifact, and the organizational context, matter more than the design file?

Figma’s bet is: yes, because all of that will come back into Figma.

It’s a bet that the canvas is the core.

And if we change the spaces where we work?

I don’t think the next dominant product workspace will look like Figma with more AI features.

I don’t think it will look like a traditional design tool at all.

More likely, an IDE with some spatial collaboration. Or a browser-based product environment where live software is directly editable, inspectable, and deployable.

It will need to involve an AI orchestration layer that sits across design systems, repos, documentation, analytics, and product management tools.

Some integration of canvas, code editor, staging environment, governance and rules system.

It’s going to look bad at first.

Early versions of what’s right are going to look worse than mature versions of the past. Awkward, incomplete, and easy to dismiss.

Figma should understand this better than most. It won the last round because the future wasn’t just a better design tool, it was a different environment for the work.

The canvas may well remain essential. The canvas-as-abstraction will not.

The canvas needs to be a place to discuss reality, not flatten it.

That is a hard, interesting problem.

What does Figma do next?

I can think of three paths.

A defensive path is to continue to expand the canvas. Build to make more and more work happen inside Figma. That will definitely produce useful features. And it might produce strong revenue. Figma is dominant, and can become stickier and more embedded.

A second path is transitional. Make the canvas more code-aware, and more interactive. Better generation and better workflows. Better import and export. This seems to be where their current moves are. It’s really useful, but it still organizes around the canvas as the product environment.

Or it might accept that the canvas - and thus Figma - won’t be the center of truth. So they build to become one of the best collaborative interfaces into that truth.

That means treating code, product context, design systems, and live behavior as the actual work. And the canvas is just a view into that. A place where teams can reason in a visual way about a system that’s already alive.

I don’t know if Figma wants to make that pivot.

Strategic change isn’t necessarily about seeing into the future. It’s about having to give up on the assumptions that make the present business work.

Multiplayer design isn’t going to go away. It still matters.

The question is where that will live when we can generate, modify, review, and ship product much closer to code.

Figma understood the last change in the center of gravity. Now that center of gravity is moving again.

I’m curious whether Figma follows it.

I’m not a neutral observer.

I’m VP of Product at Knapsack. We’re building in the place where structured design systems and product context meet AI-driven delivery.

Further reading:

• Seiz, G. & Kern, A. From Claude Code to Figma: Turning production code into editable Figma designs. Figma Blog, Feb 2026

• Banfield, R. Digital Design Isn’t Dead. It Just Got Way More Interesting. Medium, Apr 2025.

• The Innovator’s Dilemma. Wikipedia.

Article photo by Krakograff Textures on Unsplash.

It wasn’t the usual creak from the left housing. It wasn’t a thin brake pad grinding. Rory knew those sounds. Something in the drive hub.

He could feel it more than he could hear it. A hesitation in the right wheel. A second or so delay before the servo engaged.

He waited in the back room.

It smelled of stale beer. The bulb above him flickered. The plastic shade was cracked. It still cast a jaundice-yellow glow on the stained walls. He’d watched the vids of this week’s Premiership goals three times already.

Moira had said she’d be free by two. Half-past at the latest. Rory had no way of reaching her.

He’d found her through a friend. Someone who knew someone who’d worked with her at a parts plant outside Glasgow. Before it shuttered. “She’s good,” they’d said. “Skittish lass, though. You’ll have to go to her.”

This was the closest thing to compromise that she’d allowed. Meet in the back room of the Stag’s Head. The landlord was her cousin, or something, and didn’t ask questions. His chair had struggled up the slight hill to get here.

Maybe the charge regulator had been on its way out for weeks and he hadn’t noticed. He worried everything was going to fail at once.

His chair was an Akeron. Not an antique. Some subsidiary of Keida Prosthetics had bought Akeron. The sympathetic woman on the phone had told him that they didn’t honor warranties on legacy models. Official process was to go to Edinburgh, surrender the chair, wait for six weeks for a new Keida model with current firmware. And pay the equivalent of three month’s rent.

So he sat in the back room of a pub and hoped.

The door opened and a woman came in, carrying a plastic crate. The kind supermarkets used for deliveries. She was maybe fifty. Short gray hair and grimy hands. She set the crate on the table and looked at him.

“I’m Moira,” she said.

“Rory.”

She crouched and looked at his chair.

“200 model,” she said. “You said it was the right drive?”

“Something keeps clicking, and it’s slow.”

“Let me see.”

She shuffled sideways until she was facing the right side of the chair, and ran her fingers along the drive housing. Her touch was light. She grabbed a small scanner from her back pocket.

“The play in the hub’s just age,” she said. “But it’s making the motor work too hard. Or it could be the bearing seat’s worn.”

“What do you think?”

“If it’s the bearing seat, I can’t fix it here.” She was quiet for a moment. Her fingers stopped at the seam where the axle met the wheel. “But I think it’s the motor.”

She pulled open the crate. Inside, in an anti-static foam wrap, was a hub motor. Tool marks around its bolt holes. Grind marks where the serial number should be. But it was clean. New seals.

“Old 160 motor,” she said. “Same housing. Different controller firmware, but I can reflash your whole system.”

“How much?”

“Four hundred.”

“C’mon.”

Moira looked at him. “You want to know the official price?”

“I’ve not got four hundred.”

“OK. What do you have?”

He had two hundred and eighty pounds on a credit wafer. Most he’d been able to save, what with the rent and mum’s prescriptions.

He told her.

She was quiet for a minute. He listened to the pub’s holo through the wall. Someone laughed.

“Alright,” she said finally. “Two eighty. And I keep the old motor.”

“And?”

“And you keep your mouth shut about where you got it.”

“That’s it?”

“That’s it.” She pulled the crate toward her. “I’ll need an hour. You can wait in the bar.”

He couldn’t go to Edinburgh.

She wheeled him out to a worn table where he could see the holo. He shifted himself over into a wooden chair. Moira even brought over a pint. Then she pushed the Akeron toward the door.

“Moira.”

She stopped.

“Thank you.”

She didn’t turn. “I haven’t fixed it yet.”

The door closed behind her.

A story from the Static Drift universe.

Article Photo by Martin Fennema on Unsplash.

I fell into what’s probably a pretty common trap for a first year student. I read about Jeremy Bentham’s theory of Utilitarianism.

The greatest good for the greatest number.

Clean and simple and rational. It doesn’t support privilege, sentiment, private interests. It’s a simple calculation that asks what produces the most good for the most people. What could be fairer?

My tutor at Durham, the ever provocative Dr Dawson, asked me a question.

Doesn’t that potentially justify slavery?

His point was not that every utilitarian is a deliberately pro-slavery.

But if “good” is just math, then people are just variables. And abstract. So the benefit of enough people can counterbalance the suffering...degradation even...of a minority.

That’s elegant. It’s also horrific.

And we’re making systems that are getting very good at the elegant part.

If every human being on the planet could be guaranteed a good life, at the expense of one person being tortured forever, is that a reasonable bargain?

And if it is, when does it stop being reasonable? How many “good” lives are enough. A hundred? A million? Everyone?

That is not to say that reason has no place in ethics. It’s vital.

If it isn’t there, then our ethics are just instinctual. Private feelings. Tribal or national loyalty. Superstition. Power. So reason exists to help us look beyond just ourselves. Without reason we wouldn’t have the principles of consistency, fairness, and restraint.

We need it.

But not by itself.

Pure rationality has no human element.

Which is a worry about how we build AI. AI isn’t irrational. It’s not evil. It doesn’t hate us. It doesn’t want anything.

It calculates and recognizes and optimizes. It tries to reach a clean result.

And it has no body. It can’t be wounded. It doesn’t have emotion. No memory of shame. It won’t - as most people would - instinctively recoil at someone suffering.

AI never has a moment where, even if the numbers work, it would choose to say no. People do.

And it doesn’t need to hate anyone to produce inhuman outcomes. It only needs to optimize “good” when it can’t feel “bad”.

It’s not just a machine problem. We can see that lack of empathy in people, and how it applies to human policy when the people in charge narrow the frame to a binary view of efficiency.

Take foreign aid.

In a very simplistic view, the dismantling of USAID can look rational. The US saved tens of billions of dollars, and that money stayed in the US. That expenditure didn’t benefit US citizens directly, so it was a waste.

Those cuts and dismantling are estimated to cost 14 million additional deaths across the world by 2030. A third of those will be children under the age of five.

US foreign aid is a tiny share of federal spending. The total (including USAID) is about 1% of the total outlay.

Who accepts that bargain? Or, to bring it down to an individual level...how many people would you let die to save 1% of your annual budget?

The horror here is that the calculation is rational if you remove human reality. Those children are somewhere else. You don’t need to see the suffering. The savings are immediate. Those deaths - just statistics.

It’s even more disturbing in a human being. They have to refuse the empathy. But it can scale more rapidly in a machine. It can’t feel that empathy in the first place.

This is where Hannah Arendt’s phrase, “the banality of evil,” applies.

She’s talking about a moral vacancy. It’s not just that ordinary people can do terrible things. It’s that terrible things can be processed through ordinary systems. Compliance, language, bureaucracy. So the evil becomes banal, because nobody believes that they are choosing it. It’s just doing the work, even more impersonal than “just following orders.”

AI can - and if we’re not careful, will - intensify that.

An AI makes a recommendation. A dashboard says it should be accepted. A manager takes a look at that metric - it’s optimized. They approve it. A person is harmed, but it disappears into the workflow.

That’s harm as procedure. As an entirely reasonable output of a system that’s defined narrow terms and doesn’t feel the weight of what’s being done.

You don’t need to be Catholic, or religious at all, to find value in the Vatican’s recent warnings about artificial intelligence. The Pope isn’t talking about doctrine. He’s talking about dignity.

Dignity is something that doesn’t fit into the math.

I’m a person.

I’ve no doubt that I can be - and probably am - measured in various ways. My utility, productivity, and efficiency by my company. My citizenship by my nation. My purchasing power and risk score by stores and credit agencies.

But if you believe in dignity then you believe that a human can’t be reduced to those things.

It doesn’t remove tragedy. But it might at least view it as a tragedy. An understanding that scarce resource distribution creates pain, even if it can’t avoided.

A purely rational system might always justify one more acceptable loss.

The human question is different.

What can we not do? What must we not do?

AI can’t answer that for us. We struggle to answer it as a culture ourselves. I’m shocked by how much evil so many people will justify, again and again. Those same people would probably look at me as an impractical bleeding heart.

AI can help us reason. It can’t tell us when reason has to stop.

We talk about human in the loop (HITL) at work. AI systems that have human review, oversight, approval.

This is the same writ large. Human judgment, human dignity, human refusal.

This is for society as a whole.

That can only be us.

Further reading:

• Renic, N. & Schwarz, E. Inhuman-in-the-loop: AI-targeting and the Erosion of Moral Restraint. OpinioJuris, Dec 2023.

• Stonebridge, Prof L. Hannah Arendt’s lessons for our times: the banality of evil, totalitarianism and statelessness. The British Academy, Aug 2024.

• Utilitarianism - Ethics Unwrapped. UT McCombs School of Business.

Article photo by Johnny Ho on Unsplash.